**** BEGIN LOGGING AT Thu Mar 18 02:59:56 2010 Mar 18 06:57:14 http://pastebin.com/BFjBG4Ms Mar 18 06:57:20 vlc segfaults. Mar 18 06:59:06 er, sorry for the long command line. =þ Mar 18 07:00:16 interesting... mplayer is UNAVAILABLE Mar 18 07:07:57 alsaplayer works, though. once I add myself to "audio" group. Mar 18 09:10:49 hi folks Mar 18 09:13:52 * ogra gets more coffee Mar 18 09:13:55 morning asac Mar 18 09:14:55 grmbl ... /me curses having to port uboot patches to redboot ... Mar 18 09:26:16 heh, fun ... the change i should revert in redboot-imx doesnt seem to exist in our version at all Mar 18 09:30:32 hi ogra Mar 18 09:34:22 My company are about to make a product based on TI AM3517 (ARM Cortex-A8) and we're going to run linux. We are considering putting Debian or Ubuntu on it because we want a package system/distro. Mar 18 09:35:44 I forgot to say that I have a TI AM3517 ZOOM development kit Mar 18 09:38:12 I have to admit that I'm quite new to Ubuntu build-system, so you'll have to excuse me for stuptid questions Mar 18 09:40:12 nosse1: welcome Mar 18 09:40:22 The first one is linked to the choice of distro: When I have a Cortex-A8, wouldn't it then be most efficient (space and cpu) to use binaries for ARMv7? Which would implicate Lucid right? Mar 18 09:40:32 we've just started supporting OMAP3 kernels in the next release of Ubuntu Mar 18 09:41:07 nosse1: Lucid is guilty as charged Mar 18 09:43:57 I started rootstock, and when it spawns qemu, it locks up my CPU at 100%. Is this familiar? Mar 18 09:44:24 ogra <--- blame him Mar 18 09:45:12 I have no kind of feedback, so I don Mar 18 09:45:34 ...don't know if it has locked up or doing something sensible... Mar 18 09:46:38 (I'm running amd64 Karmic, and I've seen apps not working because I'm running 64-bit. Don't know if it's it, though) Mar 18 09:47:10 no, i use rootstock on 64-bit successfully Mar 18 09:48:40 BTW: Are there a seed for ubuntu which is even more minimal than ubuntu-minimal? There's a lot of apps in minimal which is not applicable in my embedded target Mar 18 09:50:19 nosse1: yes, the CPU issue is known Mar 18 09:50:24 e.g. for qemu Mar 18 09:50:28 investigation ongoing Mar 18 09:50:34 seems its triggered by apt Mar 18 09:50:39 or rather dpkg somehow Mar 18 09:50:53 asac: But is it dead or processing something? Mar 18 09:51:40 asac: Can I cat into its output somewhere? Mar 18 09:52:28 nosse1: its dead. we think its qemu being broken Mar 18 09:52:42 ogra: why is the bug not filed against rootstock nor qemu-kvm? Mar 18 09:53:49 asac: Thanks Mar 18 09:54:19 its bug 532733 Mar 18 09:54:21 Launchpad bug 532733 in qemu-kvm (Ubuntu) "apt/dpkg in qemu-system-arm hangs if a big task is installed (affects: 1)" [Medium,Incomplete] https://launchpad.net/bugs/532733 Mar 18 09:54:22 nosse1: ^ Mar 18 09:57:21 I running Karmic on my host. Would it (not for this specific thing, but generally) be better to run lucid? Mar 18 09:57:50 host system seems to make no difference afaict Mar 18 09:58:02 its lucid vm that causes this Mar 18 09:59:01 So when I'm using rootstock it also pulls down a lucid vm? I'm also pulling ubuntu-minimal and not ubuntu-netbook as noted in the bug report Mar 18 09:59:15 interesting Mar 18 09:59:22 but you are installing lucid? Mar 18 09:59:26 yes Mar 18 09:59:26 e.g. in the vm? Mar 18 09:59:36 i suspect this to be a racy thing Mar 18 09:59:43 the biggere the task is the more likely you hit it Mar 18 09:59:54 for us ubuntu-minimal works usually, but netbook triggeres this Mar 18 10:00:12 but i dontsee why ubuntu-minimal wouldnt cause this for some Mar 18 10:00:17 I can probably strace it, but I'll have to leave for a meeting first... Mar 18 10:00:18 what are your host specs? Mar 18 10:00:29 e.g. what system is this running on? Mar 18 10:00:37 wouldn't it depend on the system specs too? Mar 18 10:00:42 aah, exactly Mar 18 10:00:43 Ubuntu Karmic amd64 Mar 18 10:00:57 nosse1: your HW spec Mar 18 10:00:58 well. cpu mem etc. Mar 18 10:01:18 maybe also disk speed etc. Mar 18 10:01:41 Where can I find disk speed? Mar 18 10:02:17 hdparams? Mar 18 10:02:23 hdparm ;) Mar 18 10:02:52 Intel C2Duo T9900, 3.06GHz, 8Gb RAM, Disk: 101MB/sec Mar 18 10:02:54 well, but you could say: 5400 / 7200 / SSD etc. Mar 18 10:03:06 7200 for my case Mar 18 10:03:44 I'm sorry, but I have to leave. I'll be back. Thanks guys Mar 18 10:03:55 yeah. so from our findings it did go away if you slow down the system, so maybe your system is faster than ours (at least than mine it is) Mar 18 10:04:03 so makes some sense that you see it earlier Mar 18 10:13:02 ogra: your instructions dont work Mar 18 10:13:08 e.g. the saved qemu image isnt good Mar 18 10:13:16 getting unable to mount rootfs Mar 18 10:13:35 sure root=/dev/sda? Mar 18 10:13:57 I: Mounting temporary Image Mar 18 10:13:57 I: ARM rootfs created as /tmp/qemu/armel-rootfs-201003181055.tgz Mar 18 10:13:57 I: Qemu image saved as /tmp/qemu/qemu-armel-201003181055.img Mar 18 10:14:00 i use that img Mar 18 10:14:04 ogra: ^^ Mar 18 10:18:39 ok manually creating .img works it seems Mar 18 10:18:50 why is the kept img broken? Mar 18 10:24:06 hmm Mar 18 10:24:07 "When no option is specified QEMU uses a non privileged user mode network stack that gives the emulated machine access to the world. " Mar 18 10:24:10 that doesnt work for me Mar 18 10:24:14 ogra: ^^ Mar 18 10:24:20 i can get an ip with dhclient Mar 18 10:24:23 but not ping out Mar 18 10:24:24 or wget Mar 18 10:25:22 asac: ping is special though Mar 18 10:25:32 asac: Which network type did you use? Mar 18 10:25:39 telnet should work Mar 18 10:25:45 lool: wget doesnt work Mar 18 10:25:55 that's bad Mar 18 10:26:01 next wanted to try a tap device Mar 18 10:26:05 but hoped i didnt need that Mar 18 10:26:09 asac, what are the permissions of the img file (i havent uploaded the permission fix yet) Mar 18 10:26:13 let me run again Mar 18 10:26:13 asac: You don't Mar 18 10:26:23 ogra: permissions flag? Mar 18 10:26:27 what should be there? Mar 18 10:26:37 ogra: oh thats about rootfs not working? Mar 18 10:26:43 asac: ownership of the file needs to be the same as the user you're running the vm with Mar 18 10:26:44 well, chown the img file Mar 18 10:26:50 and u+rw obviously Mar 18 10:26:53 right Mar 18 10:27:00 let me check Mar 18 10:27:05 but i could boot it now Mar 18 10:27:24 the recent version saves the img as 666 Mar 18 10:27:27 ls -l ubuntu-arm.img Mar 18 10:27:27 -rw-r--r-- 1 asac asac 2048000000 2010-03-18 11:22 ubuntu-arm.img Mar 18 10:27:49 that should work for the asac user at least Mar 18 10:27:54 ogra: it boots Mar 18 10:28:01 i doubt that permission has to do something with network Mar 18 10:28:03 so what exactly doesnt work Mar 18 10:28:07 wget Mar 18 10:28:10 after running dhclient Mar 18 10:28:13 and getting ip ;) Mar 18 10:28:16 sudo dhclient eth0 ;) Mar 18 10:28:18 apt-get update -> errors Mar 18 10:28:26 ogra: well. read two lines above Mar 18 10:28:37 weird, works here Mar 18 10:28:51 let me check host iptables Mar 18 10:29:00 ogra: 666 uh Mar 18 10:29:10 all open Mar 18 10:29:16 644 Mar 18 10:29:29 ogra: I seriously hope it's not 666, that'd be a major security issue Mar 18 10:29:31 yeah. 666 seems overly open ;) Mar 18 10:29:36 sigh Mar 18 10:29:45 644 *doesnt boot* Mar 18 10:29:51 666 it is :-( Mar 18 10:30:07 and i have no way to make sure to know which user will use it in the end Mar 18 10:30:14 Of course you do Mar 18 10:30:19 since you removed all references to run qemu with sudo Mar 18 10:30:36 lool, how, there are many people that use rootstock with cron Mar 18 10:30:48 now network works ;) Mar 18 10:30:49 for daily local builds Mar 18 10:30:51 ogra: I've been running qemu all this time without sudo; kees setup the same thing yesterday, and didn't use sudo either or had any permissions issues Mar 18 10:30:59 * asac likes things going away ;) Mar 18 10:31:21 lool, with a 644 img owned by root.root and no initramfs ? Mar 18 10:31:40 ogra: why not chown to the user that called rootstock? Mar 18 10:31:43 oh Mar 18 10:31:54 that would be cron for people doing daily builds Mar 18 10:31:56 well, maybe rootstock should call sudo on its own to figure the real owner Mar 18 10:32:00 ogra: Why would you want to own it by root? Mar 18 10:32:23 lool, because it was loop mounterd and created as root it gets the systems default permissions Mar 18 10:32:31 wich is 644 for the creating user Mar 18 10:32:39 ogra: I personally create the image as myself, and sudo loop-mnt it Mar 18 10:32:44 there is SUDO_USER ... so yeah. not sure why its a problem Mar 18 10:32:51 you could add an option to command line for cron folks Mar 18 10:32:58 Well either use SUDO_USER or actually only call sudo when you need to Mar 18 10:33:02 so they can say: --image-user=USER Mar 18 10:33:11 that's what I typically do in my scripts; they work as root or non-root, and call sudo when needed Mar 18 10:33:25 (We actually already had this discussion) Mar 18 10:33:27 * ogra goes for asac's suggestion Mar 18 10:33:53 asac: Let's have a cp --user option too! :-) Mar 18 10:34:06 though i would prefer a more general fix ... Mar 18 10:34:19 Anyway, I filed a bug on the usage of 666; please never do that Mar 18 10:34:35 its only in bzr yet, not in the archive Mar 18 10:34:41 I filed a bug upstream Mar 18 10:34:51 fine, still, its a dev tree Mar 18 10:42:30 it all depends on the common use case Mar 18 10:42:41 i think using SUDO_USER makes sense Mar 18 10:42:59 --image-user might make sense, but you can also write a wrapper script in cron that does that i would think Mar 18 10:43:12 hmm Mar 18 10:43:20 printf ... how do i just flush stdout? Mar 18 10:44:06 * asac tries Mar 18 10:44:29 ok that worked ;)( Mar 18 10:44:31 fflush stdout Mar 18 10:44:51 Oh I thought you were asking in shell and was puzzled Mar 18 10:45:46 heh Mar 18 10:45:47 no Mar 18 10:48:25 hmm Mar 18 10:48:33 * ogra ponders how to package x-loader Mar 18 10:49:10 it sadly needs a fully configured uboot tree since the headers it relies on are generated by uboot configuration :/ Mar 18 10:50:26 punch it into uboot? ;) Mar 18 10:50:32 or make uboot produce a -dev Mar 18 10:50:43 then i would need to merge the upstream tarballs Mar 18 10:51:13 -dev seems overly complicated ... especially in the light that we will need a -dev for each uboot flavour in the future then Mar 18 10:52:01 right Mar 18 10:52:15 effectively i would like to have one uboot package based on upstream with lucid+1 (i.e. merge imx, omap and whatever else can use plain upstream or plain upstream with patches) Mar 18 10:52:36 if its soo tightly coupled upstream then it belongs together for now Mar 18 10:52:41 ogra: Really? Mar 18 10:52:47 I don't remember building xloader against uboot Mar 18 10:52:52 so i dont want to bind to closely to the architecture for the omap package now, thats supposed to be the base for the merge Mar 18 10:53:21 lool, well, i tried for zoom and beagle, both look in the boot tree for headers Mar 18 10:53:32 ogra: Which tree are you using? Mar 18 10:53:41 I just git clone gitorious:x-load-omap3/mainline.git and then: Mar 18 10:53:44 make CROSS_COMPILE=arm-none-eabi- distclean Mar 18 10:53:44 make CROSS_COMPILE=arm-none-eabi- omap3530beagle_config Mar 18 10:53:44 make CROSS_COMPILE=arm-none-eabi- Mar 18 10:53:51 and that's all Mar 18 10:54:16 I couldn't get the signed stuff to work Mar 18 10:55:01 well not the gpsign one but signGP.c worked Mar 18 10:55:06 hmm, i pulled from the zoom tree, but that should essentially be the same (just with 3630 patches added) Mar 18 10:56:52 the zoom tree expected to find ../u-boot/....*.h for building the ift image Mar 18 10:57:02 (which you need for SD) Mar 18 10:57:07 http://code.google.com/p/beagleboard/wiki/BeagleSoftCompile doesn't mention building uboot first either Mar 18 10:57:23 ogra: What are you trying to build, the MMC version? MLO? Mar 18 10:57:33 both Mar 18 10:57:40 Err that's the same thing Mar 18 10:57:44 i want a binary with a NAND and a MLO version Mar 18 10:57:49 MOL has a dos header Mar 18 10:57:58 *MLO Mar 18 10:58:01 MLO is the file which you write on your vfat Mar 18 10:58:06 right Mar 18 10:58:16 x-load.bin.ift is MLO Mar 18 10:58:23 x-load.bin is for NAND Mar 18 10:58:43 i need both Mar 18 10:59:07 and for creating the ift version the build uses the uboot header files for some reason Mar 18 10:59:24 I'm not sure what the ift version is; I don't think I used that Mar 18 10:59:28 at least in the zoom tree, i'll try the mainliune one Mar 18 10:59:48 you cp .ift to MLO and cp it to your vfat to boot Mar 18 10:59:49 Oh that's just the signed one Mar 18 11:00:07 I know what you're trying to do now Mar 18 11:00:08 its essentially a dos command.com after the signing Mar 18 11:00:15 You're trying to sign it with gpsign Mar 18 11:00:20 right Mar 18 11:00:23 Which needs u-boot's headers Mar 18 11:00:28 right :) Mar 18 11:00:29 Try signGP instead Mar 18 11:00:32 it doens't need anything Mar 18 11:00:39 and signGP didn't work for me anyway Mar 18 11:00:46 http://beagleboard.googlecode.com/files/signGP.c Mar 18 11:00:50 well gpsign did Mar 18 11:01:01 ogra: when you hang in qemu is qemu completely dead? Mar 18 11:01:05 or can you ssh into it? Mar 18 11:01:09 err gpsign didn't work for me rather, typo Mar 18 11:01:33 asac, i never tried to ssh, i use the graphical qemu version and switch to tty2 Mar 18 11:01:47 ogra: so you can still log in? Mar 18 11:01:50 yes Mar 18 11:01:58 thats how i can see top ps etc Mar 18 11:02:00 send a signal to generate a core dump then please Mar 18 11:02:09 to see the defunct dpkg --unpack passing by Mar 18 11:02:33 what do you gain from a coredump without dbgsym ? Mar 18 11:02:39 ogra: you generate a core dump Mar 18 11:02:47 then install dbgsym to analyze Mar 18 11:02:53 you can load a core dump with dbgsyms from another system Mar 18 11:02:53 right, but there are no symbols Mar 18 11:03:01 oh Mar 18 11:03:01 first generate core dump Mar 18 11:03:06 it doesn't matter, it's just a memory image Mar 18 11:03:08 then install symbols and use gdb Mar 18 11:03:27 right Mar 18 11:03:28 i'll try that later ... atm i want to get x-loader and uboot ready Mar 18 11:03:33 sure Mar 18 11:03:55 how do i start graphical qemu? Mar 18 11:04:09 use the command from rootfsfromscratch Mar 18 11:04:26 it fires up the SDL version by default Mar 18 11:04:59 i am using qemu-system-arm -M versatilepb -cpu cortex-a8 -kernel ./vmlinuz -hda arm-rootfs.img -m 256 -append "root=/dev/sda mem=256M devtmpfs.mount=0 rw" Mar 18 11:05:03 with alt-2 you get insto command mode and use send-key ctrl-alt-f2 Mar 18 11:05:04 how do i switch to tty2 ? Mar 18 11:05:10 ah Mar 18 11:05:11 ok Mar 18 11:05:23 to get back just send-key ctrl-alt-f1 Mar 18 11:05:30 alt-2 == alt gr? Mar 18 11:05:34 its annoying ... but the only way Mar 18 11:05:40 alt and 2 Mar 18 11:05:59 alt-1 gets you back to the VM Mar 18 11:07:20 ogra its alt+f2 here :) Mar 18 11:07:45 weird, alt f2 gets me the gnome exec dialog here Mar 18 11:08:01 you first have to click in the window ... so it grabs the input Mar 18 11:08:09 then alt+f2345 etc works as usual Mar 18 11:08:22 cool :) Mar 18 11:08:33 heh, i always did it through the command console Mar 18 11:08:55 because i'm annoyed if qemu grabs my mouse Mar 18 11:09:28 * asac runs bzip2 to preserve state before installing netbook Mar 18 11:10:43 lets see how big it gets Mar 18 11:10:54 heh. yeah Mar 18 11:16:31 ./signGP x-load.bin 0x40208800 Mar 18 11:16:31 make: *** [x-load.bin.ift] Error 36 Mar 18 11:16:35 bah Mar 18 11:16:57 funnily it produces x-load.bin.ift Mar 18 11:17:01 100M Mar 18 11:17:02 ;) Mar 18 11:17:06 sweet Mar 18 11:17:36 me uploads Mar 18 11:25:20 JamieBennett: all fine wfor the localization of .desktop? Mar 18 11:25:41 e..g did you get the answers from pitti? Mar 18 11:25:46 (or persia) Mar 18 11:26:10 asac: working on it. Its a lot more involved that first thought after I talked with persia Mar 18 11:26:37 and I'm learning how to do it for the first time Mar 18 11:26:41 so dh7 isnt ready for that? Mar 18 11:26:44 what a crap thing ;) Mar 18 11:26:50 asac: no Mar 18 11:27:04 I considered switching to CDBS Mar 18 11:27:06 ok Mar 18 11:27:15 dont you just have to set gettext domain ? Mar 18 11:27:16 dh7 really isnt ready? wow Mar 18 11:27:20 but I have something that should be ready tomorrow Mar 18 11:27:25 i thought LP cares for the rest Mar 18 11:27:33 ogra: we have cdbs langpack.mk Mar 18 11:27:41 seems we have no equiv. for dh7 Mar 18 11:27:59 ogra: it was a simple wrapper script Mar 18 11:28:07 langpack.mk is mostly useful to desktop packages which used cdbs so far; no pressure to port it to dh basically Mar 18 11:28:10 but it should be done Mar 18 11:28:10 now it needs a build system (setup.py) Mar 18 11:28:19 and .po support Mar 18 11:28:22 bah Mar 18 11:28:25 and gettext support Mar 18 11:28:30 e.t.c Mar 18 11:28:58 JamieBennett: which package is this? Mar 18 11:29:27 lool: webservice-office-zoho (my package) Mar 18 11:29:28 Truth is, the CDBS snippet should be a separate tool, perhaps a dh_langpack Mar 18 11:30:09 lool: I'm looking through doc, web pages, snippets of code trying to learn it too which doesn't help Mar 18 11:30:53 JamieBennett: You could copy what you need to some dh override Mar 18 11:31:14 JamieBennett: e.g. override dh_gencontrol Mar 18 11:31:27 lool: but would that be 'the proper way' ;) Mar 18 11:31:41 JamieBennett: Best way would be to add support to dh to do that automatically Mar 18 11:32:00 JamieBennett: But we have packages where we call intltool-update by hand because they are not CDBS based Mar 18 11:33:08 e.g. http://paste.ubuntu.com/397186/ Mar 18 11:34:05 perhaps not very good examples actually Mar 18 11:39:37 current cdbs does stuff in binary-predeb Mar 18 11:42:19 ogra: http://people.canonical.com/~asac/tmp/ubuntu-arm-lucid-minimal.bz2 ;) Mar 18 11:42:55 great Mar 18 11:43:05 i have several of these around though ;) Mar 18 11:43:21 then why didnt you give them to me ;) Mar 18 11:43:27 thought you didnt have that online Mar 18 11:43:48 i dont have them online Mar 18 11:44:00 and we only discussed the compressing yesterday :) Mar 18 11:45:11 ah Mar 18 11:45:16 right. so now we have it ;) Mar 18 11:45:24 yeah Mar 18 11:46:10 sigh ... i used 2G onlz Mar 18 11:46:15 thats not enough for netbook i think Mar 18 11:46:30 asac: You can grow it Mar 18 11:46:37 lool: command? Mar 18 11:46:50 asac: dd if=/dev/zero bs=1M count=200 >> lucid.img; e2fsck -f lucid.img; resize2fs -p lucid.img Mar 18 11:47:01 that's for a raw image Mar 18 11:47:09 (not sure whether you're using qcow or something) Mar 18 11:47:20 i use raw image with ext4 Mar 18 11:47:26 no part table? Mar 18 11:47:29 then the above should work Mar 18 11:47:31 let me try Mar 18 11:47:37 not much to loose Mar 18 11:47:44 rootstock doesnt create partitions Mar 18 11:47:47 Yeah, worst case you have 200MB of zeroes at the end Mar 18 11:47:56 but creates ext3 by default Mar 18 11:48:17 (or did i change that ?) Mar 18 11:48:18 ogra: i didnt use the img from rootstock, but made my own and untarred ;) Mar 18 11:49:03 ah, but you used rootstock to create the tarball ? Mar 18 11:49:21 yes Mar 18 11:49:41 ok booting with new size Mar 18 11:49:42 (i just wnt to be sure we're not to far away from the rootstock creation here when we try to debug) Mar 18 11:49:59 thought its still weird that lool never saw that issue Mar 18 11:50:28 ok it worked Mar 18 11:50:30 now have 3g Mar 18 11:50:37 that should suffice Mar 18 11:50:49 i think even 2 would for armel netbook Mar 18 11:50:54 wasnt enough Mar 18 11:50:56 (we dont have oo.o in the task) Mar 18 11:51:04 it wanted to install 1.8G after ubuntu-minimal Mar 18 11:51:11 and i only had 1.6G left or something Mar 18 11:53:19 hmm, the task never wants to install that much for me Mar 18 11:53:37 its 1835MB Mar 18 11:53:44 so less than 1.8G ... but not much Mar 18 11:53:56 ^ubuntu-netbook Mar 18 11:53:57 * ogra wonders if there is a size difference between task and metapackage Mar 18 11:54:08 doesnt the caret need to be at the end ? Mar 18 11:54:09 asac: Are you trying to reproduce the problem or just analyze the core file? Mar 18 11:54:19 lool: want to reproduce and produce core files for now Mar 18 11:54:20 yes. Mar 18 11:54:37 lool: i already reproduced it here at some point Mar 18 11:55:03 furhter above there was someone who saw this even for ubuntu-minimal Mar 18 11:55:10 i think the faster the system is the easier you get it (race) Mar 18 11:55:27 as we dont see it with dbgsym installed according to ogra Mar 18 11:55:32 right Mar 18 11:55:44 installing dbgsym just makes it pass Mar 18 11:55:54 That's odd Mar 18 11:55:58 i have never seen it in -minimal though Mar 18 11:56:22 and i havent tried -desktop since we do -netbook Mar 18 11:58:13 rcn-ee, you see the hang in rootstock too, right ? Mar 18 11:59:41 ogra, ah maybe, it's still halting at "|: Extracting zlib1g..." for me.. ;) debian qemu.. http://rcn-ee.homeip.net:81/dl/daily/ubuntu-lucid.log but it worked just fine yesterday on karmic... Mar 18 12:00:00 yes, its a lucid issue Mar 18 12:00:20 though that is something different i think Mar 18 12:00:46 yeah i was guessing that, i need to setup a karmic chroot on my server... Mar 18 12:00:54 do you have a full log too ? (the one that rootstock spits out if you ctrl-c) Mar 18 12:01:07 so you guys have fun building xloader/uboot for omap? Mar 18 12:01:19 the prob is when building lucid ... not with lucid on the host Mar 18 12:01:40 rcn-ee, well, fun is something different :) but yes, i'm fddling with that Mar 18 12:01:40 sorry, nope, it's a cron job, no terminal, and it doesn't occur when run from the command line... Mar 18 12:02:20 ah Mar 18 12:03:30 rcn-ee: I've seen the hang at zlib1g on karmic too Mar 18 12:03:36 oops, ogra ^^ Mar 18 12:03:39 well, this hang is clearly not the same one Mar 18 12:03:59 why? Mar 18 12:04:07 ogra, under what conditions is it hanging now? Mar 18 12:04:08 i thought we hang in unpack too Mar 18 12:04:10 unpacking of zlib happens in debootstrap which is run in the qemu-arm-static chroot Mar 18 12:04:44 while the task installation hangs in a VM Mar 18 12:05:06 hmm. but why wouldnt an io issue in qemu never show up in qemu? maybe its just less easy to trigger because its a fresh process for every command? Mar 18 12:05:15 and what's weird, i had the same issue when building google's x86 os where it would hang on the same file, only x86.. ;) Mar 18 12:05:16 err second qemu==qemu-static Mar 18 12:05:47 well, i have never seen it hang in karmic builds at all ... thats very new Mar 18 12:06:14 ogra: they install lucid .... didnt you say the host qemu doesnt matter? Mar 18 12:06:24 but i definately can reproduce the netbook task install hang reliably in lucid Mar 18 12:06:36 asac, right Mar 18 12:06:59 host doesnt matter Mar 18 12:07:00 yeah no luck on netbook here either.. ;) Mar 18 12:07:01 so that they see something like this in karmic qemu when trying lucid doesnt feel that off based on that Mar 18 12:07:08 but target lrelease of the build does Mar 18 12:07:18 right. armv7+thumb2 Mar 18 12:07:20 in lucid Mar 18 12:07:27 xactly my suspicion Mar 18 12:07:58 lool: why are you sure that armv7+thumb2 is working perfect in qemu? Mar 18 12:08:02 or arent you sure? Mar 18 12:08:16 asac, you mean kernel wise Mar 18 12:08:17 ? Mar 18 12:08:23 no ... qemu in general for now Mar 18 12:08:39 as rcn-ee sees it in static Mar 18 12:08:41 i dont think qemu matters here Mar 18 12:09:01 asac, but building a karmic image (unless i misunderstood) Mar 18 12:09:32 ogra: the log is named lucid from above: http://rcn-ee.homeip.net:81/dl/daily/ubuntu-lucid.log Mar 18 12:09:34 i could imagine that the verstile kernel we use doesnt properly support thumb2 Mar 18 12:09:44 asac, oh Mar 18 12:10:00 " but it worked just fine yesterday on karmic..." Mar 18 12:10:02 its probably a karmic qemu trying to run armv7/thumb2 code Mar 18 12:10:07 that confused me :) Mar 18 12:10:08 hmm Mar 18 12:10:13 rcn-ee: so what are you doing? Mar 18 12:10:21 host = karmic + target == lucid? Mar 18 12:10:28 or both karmic / lucid ? Mar 18 12:10:45 ah crap, my naming confused everyone... host debian squeeze, running latest rootstock building ubuntu-netbook for lucid target... ;) Mar 18 12:10:57 yeah Mar 18 12:10:59 debian squeeze ? Mar 18 12:11:02 so target is lucid Mar 18 12:11:03 aww Mar 18 12:11:18 thats a totally differend qemu then i think Mar 18 12:11:24 didnt lool land some arv7 stuff here? Mar 18 12:11:31 and not in debian? or are we in sync? Mar 18 12:11:38 just checking on that, it's qemu 0.11.1... Mar 18 12:11:39 well, lool usually does his qemu stuff upstream Mar 18 12:11:50 but we pull from a different upstream afaik Mar 18 12:11:53 because of kvm Mar 18 12:12:18 and the opteron server i have is too old for kvm... Mar 18 12:12:22 0.12.3-0ubuntu15 Mar 18 12:12:29 thats what we have in lucid atm Mar 18 12:12:42 but pulled from qemu-kvm since karmic Mar 18 12:13:19 rcn-ee, well, kvm doesnt gain you anything for armel emu ... but the point is that we use a different upstream Mar 18 12:13:31 rigt. thats wat busted ftas images and he lost everything ;) Mar 18 12:14:22 the question is how much difference there is Mar 18 12:14:31 especially for armel Mar 18 12:14:47 correct... i just use it to verify nightly builds.. it was working early in lucid's cycle, maybe stoped in the last two months.. otherwise i build on karmic exclusively at work.. ;) Mar 18 12:15:34 yeah, i dont test on debian ... so i dont really know if it would work or not Mar 18 12:16:05 ok hangs in iputils-tracepath here Mar 18 12:16:07 minimal should work if you build on ubuntu host for all releases we support Mar 18 12:16:14 asac, right, same here Mar 18 12:16:35 that pretty reproducable Mar 18 12:16:54 not 100% though ... there is another package it sometimes hangs ... Mar 18 12:17:03 iso-codes iirc Mar 18 12:17:17 hi, guys. Back in business! Mar 18 12:17:18 but i have only seeing it hang on either of these two yet Mar 18 12:17:25 asac: Sorry, I dont understand your question? Mar 18 12:17:45 asac: There might be qemu bugs Mar 18 12:17:50 ok Mar 18 12:17:57 it's been a toss up between iputils-tracepath and iso-codes for me.. Mar 18 12:18:21 I wonder whether a FS benchmark would show a similar issue Mar 18 12:18:22 Have you come any closer to investigating the qemu hang? Is there anything I can assist with? Mar 18 12:18:34 lool: i ran a read/write testcase here today for 1h Mar 18 12:18:39 e.g. just reading and writing all the time Mar 18 12:18:42 that didnt have issues Mar 18 12:18:47 lool, but are you sure that just telling the versatile kernel that it is a v7 CPU also enables all features properly (thumb2 etc) Mar 18 12:18:56 lool: which fs benchmark would you suggest? Mar 18 12:19:11 ok so now ... which process should i core dumb ;)? Mar 18 12:19:20 probably all dpkg/apt ones Mar 18 12:19:27 i mean i see them in cpuinfo ... but is it sure they actually work by just telling the kernel it is v7 ? Mar 18 12:19:28 not sure if i get the core from all though at the same time Mar 18 12:19:49 ogra: Uh the whole userspace works Mar 18 12:19:50 well, apt is the one that hangs Mar 18 12:20:00 ogra: All our programs are running in thumb 2 mode aren't they Mar 18 12:20:05 lool, yes Mar 18 12:20:20 so you think other SW would expose issues too ? Mar 18 12:20:22 asac: Dunno, some kernel fs benchmark to cause high IO Mar 18 12:20:36 ogra: other SW? Mar 18 12:20:43 (if it was kernel side) Mar 18 12:20:46 other apps Mar 18 12:21:07 ogra, asac: One thing you folks could try to workaround the issue are different disk emulation in qemu; e.g. cache=none, cache=writeback, etc. Mar 18 12:21:11 aren't some packages, still in arm mode? Mar 18 12:21:22 rcn-ee, very very few Mar 18 12:21:39 ogra: Well it might be a dpkg / apt bug exposed by our testing conditions or a qemu bug which triggers under high IO circunstances, perhaps a race Mar 18 12:21:44 ogra: I can't tell Mar 18 12:21:55 yeah Mar 18 12:22:16 its definitly racish Mar 18 12:22:38 ok i have a core ... not sure if it was the apt or dpkg core though Mar 18 12:22:40 i'm close to say its a race ... apt hanging in read() and the issue going away with dbgsym installed point very much to it Mar 18 12:23:30 also the I/O isnt blocked if you work from another tty Mar 18 12:23:37 while the hang occurs Mar 18 12:23:40 Is there a way I can circumvent the qemu-lockup from rootstock, so that I can build my ubuntu-minimal image? Mar 18 12:24:00 nosse1, juts ubuntu-minimal should actually work Mar 18 12:24:03 nosse1: is the hang in second stage? Mar 18 12:24:15 then try another disk emulation as lool suggested above Mar 18 12:24:32 asac: Yes. "Switching to Virtual Machine for second stage processing" Mar 18 12:24:48 nosse1, do you have a full log ? Mar 18 12:24:49 nosse1: then try to do that Mar 18 12:25:16 Another way might be to have a while sync; do sleep 1; done loop on another vt while doing the install Mar 18 12:25:28 ogra: Where can I find the log? Is it the rootstock-xxx.log file? Mar 18 12:25:54 * lool lunch & Mar 18 12:25:57 nosse1, right, rootstock should have told you it created the file when you stopped it (or it successfully finised) Mar 18 12:26:19 can you put it somewhere so i can take a look where exactly it hangs Mar 18 12:27:11 The log files doesnt actuall tell anyting.. It shows the output from a wget session. Then "I: Killed ..." Mar 18 12:27:28 whats your host os ? Mar 18 12:27:40 Karmic amd64 Mar 18 12:27:47 hmm, amd64 Mar 18 12:28:01 do you use the rootstock package from the archive ? Mar 18 12:28:05 yes Mar 18 12:28:12 and you try to build lucid ? Mar 18 12:28:17 yes Mar 18 12:28:19 ah Mar 18 12:28:31 thats fixed but not backported to karmic Mar 18 12:28:48 should I up my host to lucid then? Mar 18 12:29:34 http://bazaar.launchpad.net/~project-rootstock-developers/project-rootstock/trunk/revision/31 Mar 18 12:29:42 the rootstock scrip needs this fix Mar 18 12:30:47 (currently karmic can only build karmic) Mar 18 12:30:54 (without this fix) Mar 18 12:31:37 * ogra tries to find some food Mar 18 12:39:03 The ultimate goal of this is to throw Ubuntu into an embedded product which will go into mass production. What is the state of Ubuntu Lucid for ARM? Mar 18 12:39:21 The product is at least 12-14 mnths away from release Mar 18 12:39:51 nosse1: lucid for arm is quite good. we release beta today Mar 18 12:39:55 rootstock is busted though ;) Mar 18 12:40:16 there are a few ftbfs and thumb2 porting issues outstanding Mar 18 12:40:53 Since we're using a Zoom TI AM3517 evaluation kit as a base, is there something I can do to contribute for support for that exact kit? Mar 18 12:41:40 Pardon my silly questions, I'm quite new to the development flows of Ubuntu :o Mar 18 12:45:06 nosse1: yes, testing like you do now is good. also checking if there is software for that that isnt in the archive would help (besides the kernel/bootloader) Mar 18 12:45:23 I see from the topic that "no, we do not cross compile"? What do you mean? That all packages are built natively on some target/emu? Mar 18 12:45:36 nosse1: we build natively on real machines Mar 18 12:46:04 nosse1: if there are particular software you need to cross compile let me know which software that is Mar 18 12:46:18 currently collecting the main use cases we should support in some way Mar 18 12:47:02 nosse1: here you can see our build farm: https://edge.launchpad.net/builders Mar 18 12:47:06 we will have Qt (>=4.6) running, but if it need cross compile I'm not sure yet Mar 18 12:47:17 nosse1: is that newer than in ubuntu? Mar 18 12:47:53 than karmic, yes. I havent checked lucid yet Mar 18 12:47:55 we have Version: 4:4.6.2-0ubuntu Mar 18 12:47:56 in lucid Mar 18 12:48:02 at least thats the version for libqt4-dev Mar 18 12:48:08 so thats already precompiled in lucid for you Mar 18 12:48:13 Thats latest release from Qt Mar 18 12:48:24 if you have fixes for that that you need to get in and that are general issues we can try to get that patched Mar 18 12:48:59 * heads off to work, needs to stop playing with the xm board and send amit some usb/musb patches for omap.. Mar 18 12:49:21 how's xm? :) Mar 18 12:49:43 i like it. ;) Mar 18 12:49:58 I'm a little uncertain how the rendering system should be done in Qt and/or X11. Because the SDK to access the HW GFX accelerators are not open source Mar 18 12:50:37 nosse1: so to get unbut-minimal all you need to do is to create it from lucid (not debian) Mar 18 12:50:44 that definitly works for everyone here Mar 18 12:57:31 Yes! The patch to rootstock worked! Mar 18 13:01:12 How can I install or sign up for a builder in the build farm? Mar 18 13:02:27 I mean, the process from making a source package up until it ends up in a apt tree is really unknown to me Mar 18 13:02:50 Because we will have to make our own apt tree for the closed applications and such Mar 18 16:18:41 ogra@osiris:~/Devel/branches/omap$ sudo build-arm-chroot lucid /mnt http://192.168.2.87:9999/ubuntu-ports Mar 18 16:18:42 W: Target architecture is the same as host architecture; disabling QEMU support Mar 18 16:18:46 lool, that seems broken Mar 18 16:19:09 either drop build-arm-chroot or make it set --arch armel Mar 18 16:20:57 ogra: omap image? :-p Mar 18 16:21:00 * amitk hides Mar 18 16:21:14 * ogra hits amitk with a rootstock Mar 18 16:21:29 :) Mar 18 16:22:09 :) Mar 18 16:22:21 the kernel should the archive tomorrow Mar 18 16:22:30 *hit the archive Mar 18 16:22:45 great, i'll make sure x-loader and uboot are uploaded before end of day tomorrow Mar 18 16:23:02 and will work on cdimage scripts on the weekend Mar 18 16:25:26 you guys doing omap support now? Mar 18 16:30:40 OOooh. Mar 18 16:31:24 heh, every time I try to compile a "newer" u-boot (u-boot-omap) for my beagleboard, I find it mistakenly detects it as Board Revision Ax/Bx (instead of C4, as it should be). Mar 18 16:31:50 hmm, what benefit do a newer u-boot and/or x-loader give? Mar 18 16:33:10 http://pastebin.com/BFjBG4Ms -- vlc segfaults. Mar 18 16:33:11 er, sorry for the long command line. =þ Oh, and mplayer is UNAVAILABLE. Mar 18 16:34:10 heh, I left in my sd card with Angstrom (ew) on it, and grub on my host "Found unknown Linux distribution on /dev/mmcblk0p2 Mar 18 16:37:26 DanaG, mplayer is availabel in multiverse Mar 18 16:38:00 hmm, I'll check sources.list. Mar 18 16:38:14 for vlc, please file a bug Mar 18 16:38:42 (and hope that someone from the community puts some work into it) Mar 18 16:39:10 How do I trigger apport via CLI? Mar 18 16:39:34 ubuntu-bug vlc Mar 18 16:39:40 should work Mar 18 16:39:42 Will it get the gdb output? Mar 18 16:40:02 if there was a crash file file created it will just attach it Mar 18 16:40:17 Cool. Mar 18 16:40:24 check /var/crash/ Mar 18 16:41:25 nope, no vlc .crash file. Mar 18 16:43:27 also, when trying to run amttool on the beagle: Mar 18 16:43:28 Can't locate SOAP/Lite.pm in @INC (@INC contains: /etc/perl /usr/local/lib/perl/5.10.1 /usr/local/share/perl/5.10.1 /usr/lib/perl5 /usr/share/perl5 /usr/lib/perl/5.10 /usr/share/perl/5.10 /usr/local/lib/site_perl .) at /usr/bin/amttool line 4. Mar 18 16:43:29 BEGIN failed--compilation aborted at /usr/bin/amttool line 4. Mar 18 16:44:44 looks like a missing dependency ... Mar 18 16:44:48 a bug as well Mar 18 16:45:13 the community support for apps in universe under armel isnt that great yet Mar 18 16:45:23 (any help appreciated :) ) Mar 18 16:46:09 hah, nvidia-kernel-common exists on ARM restricted. Mar 18 16:54:06 heh, and so do things like virtualbox-guest-additions. Mar 18 17:04:34 yo Mar 18 17:09:46 weird... ssh into beagleboard and run mplayer... it outputs LOCALLY to pulseaudio on the ssh client. Mar 18 17:10:54 ah, had to un-export DISPLAY= Mar 18 17:12:48 ogra: build-arm-chroot does set arch == armel Mar 18 17:13:01 lool, didnt for me Mar 18 17:13:01 ogra: But you're running this on arm, right? Mar 18 17:13:04 no Mar 18 17:13:18 i just tried to quickly build an arm chroot Mar 18 17:13:18 * DanaG wishes he had a spare DVI display. Mar 18 17:13:22 ogra: Ah I know what happened Mar 18 17:13:26 Right now, my beagleboard is entirely headless. Mar 18 17:13:42 lool, it also doesnt throw the deprecation warning Mar 18 17:15:07 ogra: uploaded Mar 18 17:15:22 ogra: http://paste.ubuntu.com/397356/ Mar 18 17:16:16 lool, hmm Mar 18 17:16:32 lool, i didnt run build-arm-chroot with a path Mar 18 17:16:41 ogra: sudo added one Mar 18 17:17:06 oh Mar 18 17:17:14 indeed :) Mar 18 17:17:24 yo Mar 18 17:17:55 ghurt Mar 18 17:18:21 juje Mar 18 17:19:25 http://blokker1999.info/andromnia/README.txt Mar 18 17:19:33 dis ewe do? Mar 18 17:21:20 ewe no talk? Mar 18 17:25:09 KingMuty, try #andromnia Mar 18 17:25:29 ok thx Mar 18 17:29:40 dey iz dead! :'( Mar 18 17:47:36 interesting... so, if I wanted to have pulseaudio run as my local user... how would I make it auto-start on a headless thingy? Mar 18 17:48:55 now, here's something funny to do: Mar 18 17:49:09 screen /dev/ttyUSB0 115200 on host, to beagleboard serial port. Mar 18 17:49:22 now run byobu within that. Mar 18 17:49:32 And then within byobu... use amtterm to go back to the host. Mar 18 17:49:41 ... and then run another byobu there. Mar 18 17:50:43 or even better... make the first "screen" a byobu, too. **** BEGIN LOGGING AT Thu Mar 18 22:39:53 2010 Mar 19 00:45:30 oh yeah, so are we going to have an omap kernel in the repos? Mar 19 00:46:33 argh, not enough free space on my sd card for a whole kernel tree. Mar 19 00:46:44 And no cross-compiler in the amd64 host's repos. Mar 19 00:47:46 yeap, it looks like it.. just emailed amit the usb config changes. ;) Mar 19 00:47:59 oh yeah, I hope it'll have the usb-audio gadget, and such. =þ Mar 19 00:48:08 the rcn-ee kernel doesn't have those. Mar 19 00:48:29 laughs, what's the config change, and it will. ;) Mar 19 00:49:18 here are the lines I see relevant in the config right now: Mar 19 00:49:52 CONFIG_USB_ETH=y CONFIG_USB_ETH_RNDIS=y CONFIG_USB_ETH_EEM=y # CONFIG_USB_GADGETFS is not set # CONFIG_USB_FILE_STORAGE is not set # CONFIG_USB_MASS_STORAGE is not set # CONFIG_USB_G_SERIAL is not set # CONFIG_USB_MIDI_GADGET is not set # CONFIG_USB_G_PRINTER is not set # CONFIG_USB_CDC_COMPOSITE is not set # CONFIG_USB_G_MULTI is not set CONFIG_USB_OTG_UT Mar 19 00:50:07 Which ones to enable, is partly subjective. Mar 19 00:50:52 oh, and the audio one doesn't seem to be there. Mar 19 00:51:04 time for some googling. Mar 19 00:52:20 http://docs.blackfin.uclinux.org/doku.php?id=linux-kernel:usb-gadget:audio Mar 19 00:52:24 those can be tricky.. if you look in my 2.6-stable repo, under patches/rcn/ there is a 002-defconfig-disable-ga***.diff, basicly you can break usb based rootfs on bx and c2 boards.. ;) Mar 19 00:52:56 sd card based rootfs don't care. .;) Mar 19 00:53:22 hmm, which repo should I clone, to go experimenting with the stuff? Mar 19 00:53:49 I'm curious how that file gadget thingy works.... what does the host see... a raw disk, or some magic files? Mar 19 00:54:31 2.6-stable is 2.6.32, 2.6-dev is 2.6.33, 2.6-mailine is broken (2.6.34-rc1) and lucid-ti-omap is what ubuntu will have. Mar 19 00:54:53 oh, and one issue I do notice: [10468.479431] Class driver suspend failed for cpu0 Mar 19 00:55:48 hmm, I do see... allowing raw disk access from both host system and the ARM device... sounds like a way to have Very Bad Things happen. Mar 19 00:55:52 =þ Mar 19 00:56:24 oh, I see... it's file-backed, not raw-disk-backed. Mar 19 00:56:28 yeah it does, i did that on an 8-bit usb micro to see what would happen.; Mar 19 00:57:30 8-bit usb micro? micro-sd card? Mar 19 00:58:08 http://www.linux-usb.org/gadget/ Mar 19 00:58:17 Oooh, u-boot environment vars for MAC address and such. Mar 19 00:58:19 oh, it was an avr 8bit usb micro, was messing around with usb file systems.. Mar 19 00:58:32 er, wait, no, they just go to cmdline. =þ Mar 19 00:59:46 I'm curious how the audio gadget would work... playback to device from host would appear as an input stream on the guest, I suppose. Mar 19 01:00:20 oh yeah, and for rndis in win7 64-bit, you can just manually choose "RNDIS-Compatible" in device manager -- and it's even signed. Mar 19 01:00:28 i'm guessing it would be just like those cheap usb 5.1 audio plugins, the beagle would play the stream.. Mar 19 01:00:45 I have one of those USB sound cards (CM106). Mar 19 01:00:49 It does some rather weird stuff. Mar 19 01:00:54 http://pulseaudio.org/ticket/678 Mar 19 01:02:11 also weird: there's no kernel here -- http://rcn-ee.net/deb/kernel/beagle/lucid/v2.6.33.1-dl.2/ Mar 19 01:02:50 yeap, i messed up the script last week and didn't clear the queye.. ;) Mar 19 01:03:21 I wish the beagleboard had more audio outputs. Mar 19 01:03:54 ... or would at least hide the non-relevant stuff in the hideously complicated mixer. =þ Mar 19 01:04:01 when one of those missing stuff, just take a look at: http://rcn-ee.homeip.net:81/dl/farm/log/ shows the status of my farm.. ;) Mar 19 01:04:41 basicly the *.deb file wasn't spelled exactly as i was expecting.. . Mar 19 01:05:14 oh, and ureadahead main process terminates with status 5 -- happens on my host system, too (drm-next kernel). Does ureadahead need some non-upstream patches? Mar 19 01:05:30 it's chewing on that one right now.. http://rcn-ee.homeip.net:81/dl/farm/log/BUILDING-2.6.33.1-dl2_1.0-lucid.txt should be done in another 4-5 hours... Mar 19 01:05:40 yeah it does... Mar 19 01:06:16 It also breaks the plymouth splash on my host. And makes it slow to log in. Mar 19 01:06:47 oh yeah, and thank you greatly for making those kernel and rootfs images available -- it's a great help for using the board. Mar 19 01:06:55 or i thought it did... http://kernel.ubuntu.com/git?p=ubuntu/ubuntu-lucid.git;a=shortlog;h=refs/heads/ti-omap i'm going blind... Mar 19 01:07:30 also funny on ARM: things like nvidia-kernel-common and virtualbox-guest-additions still exist. Mar 19 01:07:56 i've had reports that "console-setup-mini" greatly speeds up serial prompt login... Mar 19 01:08:37 I've come to have a great appreciation for serial console -- my host laptop has serial-over-lan; awesome for grabbing stacktraces. Mar 19 01:08:54 yeah there's a lot of funny ones, libdrm_intel/radeon/etc.. plymouth actually kinda worked for me today on bootup... although it didn't go any farther.. Mar 19 01:08:58 I got a kernel panic on my other laptop (a netbook)... and had NO way whatsoever to get a full stacktrace. Mar 19 01:09:14 I'm using drm-next on the host, for the sake of radeon KMS power-management. Mar 19 01:09:52 me too, i'm torn between getting the xm beagle working or test agd5x's new pm patches.. ;) Mar 19 01:10:19 I'll wait for it to go into the kernel-ppa stuff. Mar 19 01:10:36 oh yeah, does xm have gigabit ethernet, or just 10/100? Mar 19 01:11:22 not sure, it's smsc and i'm guessing 10/100.. Mar 19 01:11:38 lan9514 Mar 19 01:12:16 cool, it's a usbhub/usbethernet.. so it should be faster then micrel's zippy2... Mar 19 01:12:44 SPI ethernet strikes me as weird. Mar 19 01:13:15 it is... but it's perfect for ethernet enabling a pic 8/16bit... Mar 19 01:13:27 That 3-port-hub-with-ethernet that specialcomp.com sells is pretty cool -- it's also what the macbook air should have come with. =þ Mar 19 01:14:23 oh yeah, do you have the monochrome LCD from specialcomp.com? I couldn't quite figure out how you'd tell the kernel to use 480x272. Mar 19 01:14:57 nope i don't.. usually headless installs.. are you connecting between the lcd pins or hdmi connector? Mar 19 01:15:23 er, a classmate has it (and is going to try it this weekend); it goes to the lcd-controller pins and expansion connector. Mar 19 01:17:26 not sure.. it's the one thing i haven't spent enough time trying to do a one of these boards.. Mar 19 01:18:01 what I've done with mine so far: mostly just mess around with pulseaudio. It works awesomely well even over usb-net. Mar 19 01:18:35 yeap, that was too easy.. xm still locks.. Mar 19 01:19:32 Do you talk to the people who do the kernel-ppa stuff? I've noticed that it has all of STAGING disabled, as well as CONFIG_MMC_RICOH_MMC. Mar 19 01:20:42 I think they use the same config as the offical release, so what ever was enabled, gets enabled... Mar 19 01:22:40 .config:2069:warning: symbol value 'm' invalid for MMC_RICOH_MMC .config:4418:warning: override: reassigning to symbol STAGING **** ENDING LOGGING AT Fri Mar 19 02:59:56 2010