[02:32] <Neko> anyone know of any progress with bug 628587 ?
[02:32] <ubot2> Launchpad bug 628587 in ubiquity (Ubuntu) "Ubiquity 2.3.13 (oem-config) fails to create initial user (affects: 1) (heat: 234)" [Undecided,New] https://launchpad.net/bugs/628587
[02:40]  * persia looks
[02:45] <persia> Looks to me like rcn-ee is the only person chasing that report.  If I'm understanding the bug correctly, something odd is happeng when user-setup is called.
[02:46] <persia> Could either do additional bisection by bzr revisions from lp:ubiquity OR turn on all the debugging output and try to pinpoint the offending code that way.
[02:46] <persia> rootstock images tend to be weakly supported (if it's an issue with rootstock itself, the rootstock devs are typically happy to fix it: if it's an issue in Ubuntu, most folk are willing to review a patch, but only support debian-cd derived images)
[03:36] <Neko> persia, indeed but right now I'm having trouble getting even a minimal image, which I then want to use to set up livecd-rootfs stuff like you suggested
[03:37] <Neko> it's really annoying :(
[03:37] <Neko> setting the login details is okay I guess
[03:37] <Neko> it just adds extra steps for everyone involved
[03:38] <persia> Given that lucid works, you might try creating a lucid environment, and using that to run debian-cd and livecd-rootfs
[03:43] <Neko> persia, doing that now :D
[08:21] <hrw> morning
[08:41] <ndec> ogra: hi!
[08:41] <ogra> hey
[08:41] <ogra> how was your release ?
[08:41] <ndec> ogra: how can I know which version of MLO is put in an image?
[08:41] <ogra> i guess only from the serial console
[08:42] <ndec> release was bad ;-) main problem is our non standard gst packages... that cause headache to install...
[08:42] <ogra> since MLO doesnt get installed as a package (we only use the binary in the boot partition)
[08:42] <ndec> add to play with pinning...
[08:42] <ogra> ouch
[08:42] <ndec> well, somehow it would be nice to track (in the manifest?) which source for uboot and MLO is used in image..
[08:43] <ogra> the manifest only holds installed packages
[08:43] <ogra> the image will always have the version that was recent in the archive when the iage was built ...
[08:44] <ogra> https://edge.launchpad.net/ubuntu/maverick/+source/x-loader-omap4/L24.9git20100901-0ubuntu1 is in the current ones and i dont expect that to change unless we add ES2.1
[08:45] <persia> When we do a proper release, it becomes easier to track, as the archive is frozen, and another created.
[08:45] <ogra> u-boot has a bit less obvious version number https://launchpad.net/ubuntu/maverick/+source/u-boot-linaro/2010.09~rc1.1-0ubuntu2 (we need to talk to linaro if you need that changed)
[08:47] <persia> Well, depends.  Seems we changed it at least once since the last upstream.
[08:47] <persia> (but definitely *best* to talk to linaro: forks are to be limited)
[08:47] <ogra> but you dont know from when the upstream checkuot it
[08:48] <ogra> i'm not sure the tag matters at all since they merge multiple branches
[08:48] <persia> Far as I'm concerned, linaro is the upstream for that.
[08:48] <persia> I know there's groups upstream from linaro, but they are outside the scope of my attention :)
[08:49] <ogra> doesnt help much if you want to know about one specific branch
[08:49] <persia> Yeah.
[08:49] <ogra> i.e. omap4
[08:51]  * ogra hopes todays image resizes again
[08:51] <persia> Anyone know anything about tbb?  I see http://software.intel.com/en-us/forums/showthread.php?t=67133 but I'm having trouble determining if ARM support is present in the Ubuntu version.
[08:51] <ogra> there arer days where i really ponder to drop set -e from the scripts
[08:52] <persia> Please don't.  Better to find out when there is an issue.
[08:52] <ogra> well ...
[08:53]  * persia has spent the majority of the past week fixing up type naming for a package that just made it work, rather than doing it right
[08:53] <ogra> (indeed i would never drop it, but it costs a day usually)
[08:53] <persia> Better a day here and there than a week or two for some poor sod in six months.
[08:53] <ogra> especially for such minor changes as the one that broke yesterdays image
[08:54] <ogra> colin hardcoded linking mtab to /proc/mounts in initramfs-tools/init ... which we do since day one
[08:54] <ogra> (in jasper)
[08:55] <ogra> so the script died silently and just moved on without resizing
[08:55] <persia> That's probably just a coordination issue.  We need better means of tracking changes on surrounding packages of interest.
[08:55] <ogra> well, we have the changelog ...
[08:55] <persia> Relying on people to contact other people isn't a good way to do it, and things change fast enough that it's not possible to read the patches anymore.
[08:56] <persia> Sure, but who reads all the changelogs anymore?  I used to do it, but I often can't these days, as they come too fast.
[08:56] <ogra> bah, and the LED heartbeat is way to fast
[08:56] <ogra> XM at least has a 1sec interval
[08:57] <ogra> ah, great, it is resizing again
[08:59]  * ogra now has a hectical blinking board in front of him
[08:59] <ogra> i think the frequency is 25-30ms or so
[09:00] <ogra> sigh and the swap creation is really taking long
[09:28] <ogra> gah, and indeed apturl doesnt work as expected :(
[09:32] <hrw> Xm, panda... /me -> picocom ttyS2 to check beagleboard c3
[09:33] <hrw> ops.. its ttyS6
[09:38] <fgu> ogra: hi just a question about your toshiba ac-100...you succeeded to get root access etc on it?
[09:38] <ogra> fgu, yes, i managed to brick it too :P
[09:39] <fgu> ogra: gloups...because I have just found this one: http://tosh-ac100.wetpaint.com/
[09:39] <fgu> coming from here: http://forums.computers.toshiba-europe.com/forums/thread.jspa?messageID=217864&#217864
[09:41] <hrw> ac100 looks nice of photos
[09:42] <ndec> ogra: about the images, is the latest (20109020) okay?
[09:42] <fgu> hrw: hi, yes but regarding the test for video playback, some people are finding it too short in battery life...
[09:43] <hrw> fgu: for battery life I have my x86-64 laptop
[09:43] <fgu> hrw: ;o)
[09:44] <hrw> fgu: CULV intel core2duo with 8-10h life
[09:44] <ogra> ndec, yes, PPA install is broken still (working on that) and swap file creation adds a few mins to the resizing process
[09:44] <fgu> hrw: real 10h ??
[09:44] <ogra> ndec, beyond that it should be good to use
[09:45] <hrw> fgu: will test it during trip to uds-n. so far never used it for longer then 5-6h
[09:46] <fgu> hrw: sounds more realistic..but that's nice already if it's a real 6h
[09:48] <hrw> booting bbC3 is fun. local-premount starts, then nothing for long long time then oom, then normal boot
[09:48] <hrw> I really need to upgrade ;D
[09:50] <hrw> "apt-get -f -yes install", then "apt-get install aptitude", then remove all x11 related stuff
[09:50] <ogra> fgu, the rooting works fine ... and if you are not as silly as me to link /bins/sh to something broken, you can even boot afterwards :P
[09:51] <ogra> fgu, i had an ubuntu chroot running on it just fine and also a loopback vnc session to an ubuntu desktop
[09:51] <ogra> (just very slow through vnc)
[09:51] <ogra> but maverick runs fine in that setup
[09:52]  * ogra will return the device today and see if he can get an replacement or if they send it in to toshiba
[09:57] <hrw> ~curse sd slot on omap3 and its lack of any speed
[09:57] <fgu> ogra: good job even if the end of the story is a brick netbook...hope you will get a new one then...
[09:58]  * ogra too :)
[09:58] <fgu> ogra: my previous links were droid related..so at the end not usefull for you I think..
[09:59] <ogra> rageagainstthecage-arm5.bin works fine for rooting
[09:59] <fgu> ogra: yep, ok, I have read this one
[09:59] <ogra> then creating a passwd file with root user and making busybox suid root gets you permamnent root
[10:00] <fgu> ogra: well done!
[10:00] <ogra> still that doesnt get you any bootloader access though
[10:00] <ndec> ogra: about the PPA install, so is done exactly? just the setup of the PPA, or do you install some packages?
[10:00] <ogra> which is essential for changing the android params to a sane OS
[10:01] <hrw> ogra: replace kernel with uboot+kernel image
[10:01] <ogra> ndec, there is an icon on the desktop (with the next build, i messed something up in the current one) if you click it, it is supposed to fire up software-center with the offer to enable the ppa and install
[10:01] <ogra> hrw, where ?
[10:01] <ogra> hrw, there is no place you can find the kernel
[10:01] <ndec> ogra: thx.
[10:02] <hrw> ogra: auch.. no flash partition even?
[10:02] <ogra> hrw, its nvidia crap, it uses some hidden NAND to hold bootloader, kernel and initrd
[10:02] <hrw> ok, no question asked then
[10:02] <ogra> even on the tegra dev boards you can only use nvflash to write them
[10:02] <ogra> but that requires a special flash7recovery mode which was patched out in the toshiba version
[10:03] <persia> !ohmy | ogra
[10:03] <ubot2> ogra: Please remember that all Ubuntu IRC channels share the same attitude of providing friendly and polite interaction with all users of all ages and cultures. Basically, this means no foul language and no abuse towards others.
[10:03] <ogra> all that bootloader can do is install update.zip files for android
[10:03] <ogra> yippiee ! i managed to trigger an ohmy from persia :)
[10:04] <ogra> (or do a factory reset)
[10:04] <persia> Just hard to get folks to cooperate when you insult their product scatalogically.
[10:04] <ogra> persia, yeah, i'm just upset by the setup
[10:04] <ogra> (funny sentence ... would make a good T-Shirt print)
[10:05] <persia> I understand.  Even more so because it's useless to you unless it runs the OS you use, and it became even more useless in the process :(
[10:06]  * ogra takes a coffee break
[10:31] <hrw> arm people: fpc fails to build, hedgewars also fails. both require fpc to be present. any way to solve?
[10:33] <persia> hrw, fpc needs bootstrapping.  It's been done N times, but never actually works at the moment lamont tries to do it in the archive.
[10:33] <persia> Figuring out precisely what goes wrong for lamont (when it works for other folks), and ensuring that is fixed is the trick.
[10:33]  * persia looks up the bug report
[10:33] <hrw> is same with haskell?
[10:33] <suihkulokki> fpc bootstraps nice these days, you can ofcourse shortcut that by taking the fpc from debian
[10:34] <persia> haskell is broken for other reasons.
[10:34] <persia> suihkulokki, We're shortcutting, but it still doesn't work for lamont.
[10:34] <persia> hrw, For haskell, I'd recommend asking Laney in #ubuntu-motu: he's been driving that mostly.  He'd really appreciate some help testing some things (he has no arm hardware today)
[10:35] <persia> Right, fpc is bug #67544
[10:35] <ubot2> Launchpad bug 67544 in fpc (Debian) (and 3 other projects) "Bootstrapping needed for fpc for armel (heat: 12)" [Unknown,Fix released] https://launchpad.net/bugs/67544
[10:36] <persia> Fixing fpc would mean we get something like 20 packages that we don't have today: the trick is to get something that works for lamont: just because the rest of us can do it doesn't mean it can be uploaded.
[10:39] <hrw> persia: need to get my bb/c3 upgraded - will take most of day probably
[10:39] <hrw> die sd card, die
[10:40] <persia> hrw, Won't help.  Fixing it requires building it on a bbg3.0 with a buildd chroot image.
[10:41] <persia> buildd chroot images can be downloaded from launchpad.  Getting a bbg3.0 running the config on the buildd is a bit trickier (not least because one has to have the right hardware first)
[10:41] <hrw> persia: I prefer to do builds on local hardware before going to externals
[10:42] <persia> hrw, I understand.  I'll tell you now that if you install maverick on your bb/c3, and you download the source, and you grab the bootstrap binaries (either doko's or from Debian), you will bootstrap it successfully.
[10:42] <persia> The issue is only in finding out why it doesn't work on the buildd, and getting that working, and uploaded.
[10:43] <persia> The actual code is fine.
[10:43] <hrw> ok, thx
[10:43]  * persia has bootstrapped it twice on armel, with perfect success, and no archive results :(
[12:45] <cjjnjust> hello anyone online
[12:50] <persia> Lots of folks.  /names will give you a list :)
[13:10] <rsalveti> morning
[13:10] <rsalveti> hm, seems we have images now
[13:45]  * ogra_ac waves from a new ac100
[13:45] <dmart> ogra_ac: android, or ubuntu?
[13:46] <ogra_ac> dmart android, i managed to root it but bricked it yesterday
[13:46] <persia> ogra, The RMA worked?  Congratulations!
[13:46] <ogra_ac> yeah, just play dumb and they feel pity for you
[13:46] <dmart> ogra_ac: sounds more interesting, are there instructions somewhere?
[13:48] <ogra_ac> dmart someone posted them today above, essentially you just need a terminal app and run a binary called rageinthecage or something like that  ... next term you open is rooted
[13:48]  * ogra_ac will root this one too later today ... but this time not delete /bin/sh :P
[13:49] <ogra_ac> its really embarassing
[13:50] <ogra_ac> dmart, btw, ubuntu maverick chroot works fine, i also had a vnc loopback connection to a localhost chroot that worked buut wasnt of much use (to slow)
[13:50] <ogra_ac> but i had a desktop running
[14:04] <dmart> ogra_ac: heh :)
[14:04]  * dmart has had similar fun replacing /sbin/init with garbage in the past
[14:05] <ogra_ac> at least i found out that the recovery mode works in two parts and that the second part requires a shell :)
[14:06] <dmart> though not as embarrassing as hexedit /dev/tty2 <>/dev/sda >&0 2>&0 (I was having a bad day that time)
[14:16] <ogra_ac> ouch
[14:54]  * robclark discovers sudo apt-get remove maximus
[14:54] <ogra> heh
[14:54] <robclark> makes gnome session behave much more like gnome session
[14:54] <ogra> for fixing the ugly fullscreen bug ?
[14:54] <robclark> yeah
[14:54] <robclark> I wonder why maximus starts for gnome desktop session?
[14:55] <ogra> its easier than removing it ... you can just disable it in the gnome session
[14:55] <ogra> in the session properties
[14:55] <robclark> ahh, ok.. where are session properties?
[15:01] <ndec> robclark: system/preferences/startup applications
[15:01] <robclark> ahh.. gui even.. I was assuming some config file hidden somewhere ;-)
[15:04] <ndec> robclark: if you prefer, there should be a config file hidden somewhere!
[15:04] <robclark> ndec: heheh.. no, it's ok..  I was just not looking for the easy solution ;-)
[15:06] <rsalveti> or at least gconf-like cmd line
[15:15] <jo-erlend> I have an IGEPv2 board. I've installed Maverick on a memory card, and it seems to boot. However, the screen goes black when X starts up. Any ideas on how to fix this?
[15:24] <armin76> replace with omap4 :D
[15:25] <hrw> ;D
[15:25] <hrw> first they have to appear on market
[15:26] <jo-erlend> heh. I think the hardware is quite sufficient for my needs, but it would be nice if I could see what I'm doing. :)
[15:26] <robclark> jo-erlend: anything suspicious in /var/log/Xorg.0.log?
[15:26] <ogra> jo-erlend, i think the linaro folks have such HW, though i'm not sure they use actual maverick images on them
[15:27] <jo-erlend> robclark, you're smart, you are! I'll have a look.
[15:31] <jo-erlend> robclark, indeed, it does. Perhaps you could have a look and see if you get any good ideas?
[15:31] <robclark> jo-erlend: can you pastebin?
[15:33] <jo-erlend> robclark, http://pastebin.com/6hAjUwCW
[15:33]  * robclark looks
[15:33] <jo-erlend> it looks to me like it doesn't detect any screens.
[15:34] <robclark> hmm, no framebuffers?
[15:34] <robclark> ls /dev/fb*
[15:34] <jo-erlend> I have no way of doing that, since I haven't got either network or screens to see what the results are.
[15:34] <robclark> consolefb would still work with no fb device files.. but x wouldn't
[15:34] <robclark> oh... doh
[15:35] <robclark> hmm
[15:35] <jo-erlend> perhaps I can deactivate X in order to get a console and then see what ls /dev/fb* gives me? How do I do that?
[15:35] <robclark> add the word "text" to your bootargs
[15:35] <rsalveti> weird thing is that it seems you don't have any fb device
[15:36] <rsalveti> are you able to get anything before X starts?
[15:36] <jo-erlend> robclark, in what file do I do that?
[15:36] <ogra> boot.scr or in u-boot prompt
[15:36] <jo-erlend> rsalveti, well.. First the screen is completely white, then it becomes a psychedelic mess of colours, and then it goes black.
[15:37] <ogra> sounds like my touchbook
[15:38] <jo-erlend> ogra, I don't get any prompt, and I have no boot.scr file. The install guide told me to remove all files except uImage.
[15:38] <ogra> you should have an u-boot ptrompt on the serial console
[15:39] <ogra> *prompt
[15:39] <jo-erlend> I have no serial console :(
[15:39] <rsalveti> ouch
[15:39] <rsalveti> no network, no screen and no console :-)
[15:39] <jo-erlend> right. :)
[15:39] <ogra> thats bad indeed
[15:39] <rsalveti> but you can change it at your sd card
[15:39] <jo-erlend> change what?
[15:40] <rsalveti> but to debug it properly you should at least have access to the serial console
[15:40] <rsalveti> boot.scr
[15:40] <jo-erlend> I have no such file.
[15:45] <ogra> you had it initially, right ?
[15:45] <ogra> (since you said you had to delete everything)
[15:45] <ogra> what howto were you following btw ?
[15:45] <jo-erlend> ogra, http://labs.igep.es/index.php/How_to_get_the_Ubuntu_distribution#Ubuntu_10.10_.28Maverick_Meerkat_BETA.29
[15:45] <jo-erlend> yes, I think I had it initially.
[15:46] <ogra> lol
[15:46] <jo-erlend> I can give it another go and skip deleting those files.
[15:46] <ogra> why do they call it boot,ini ?
[15:46] <jo-erlend> I have no idea.
[15:46] <ogra> boot.ini is whats boor.scr on all other omap systems
[15:47] <rlameiro> jackd doesnt start on maverick, it hangs in there i thing crashing, how do i collect data of it runnig to give it to bug hunters?
[15:47] <ogra> so you want to edit the boot.source file and re-run the mkimage command
[15:47] <ogra> but given that you have no output at all, adding "text" to the bootargs wont get you much further i fear
[15:47] <jo-erlend> oh, ok. Thanks.
[15:48] <ogra> rlameiro, best is to have network and run "ubuntu-bug jackd"
[15:48] <ogra> that will collect all info
[15:49] <rlameiro> yea , but i will want to collect some more data before that, is it possible?
[15:49] <ogra> then you need gdb i guess
[15:49] <rlameiro> will it run jackd to see what is going?
[15:49] <ogra> no
[15:49] <ogra> it will collect a defined set of logs and info thats defined in the jackd package
[15:49] <rlameiro> i want to give maximum info possible
[15:50] <ogra> fi you run a graphical session, apport should come up and collect a crashdump
[15:51] <ogra> non graphically you can take a look in /var/crash ... that might have a coredump file for the crash, but be careful, it contains your RAM content (might have passowrds etc)
[15:51] <rlameiro> problem is that it doesnt crash, it hangs, i do a ps and the process is active
[15:51] <ogra> well, then gdb and capture a backtrace
[15:51] <rlameiro> lol, i can tell my pass :D 123456
[15:51] <jo-erlend> ogra, it might be imagination, but the colorful screen seemed to stay longer. The screen still went black though.
[15:51] <rlameiro> so i run gdb <COMMAND>
[15:51] <ogra> https://wiki.ubuntu.com/Backtrace
[15:52] <rlameiro> ogra thanks
[15:58] <rsalveti> ogra: jo-erlend: I guess by removing the uInitrd the sd card will not be resized
[15:58] <jo-erlend> rsalveti, what does that mean?
[15:58] <rsalveti> from the pre-installed image
[15:58] <rlameiro> lol gdb crashed my board
[15:58] <ogra> rsalveti, well, by following these instructions the image will also stay completely unconfigured
[15:59] <rsalveti> yep
[15:59] <ogra> it wont run oem-config ever
[15:59] <rsalveti> a really ugly hack :-)
[15:59] <ogra> these instructions are quite a mess
[15:59] <jo-erlend> any computer shop should have what I need for a serial connection to the board, right?
[16:00] <rsalveti> what kind of connector do you have at your board? is it similar to the beagle one?
[16:01] <ogra> jo-erlend, USB to serial adapter ... and probably a gender changer for the sub-d port or a null modem cable
[16:01] <hrw> igep uses icd10 connector
[16:01]  * ogra has no idea how the IEGP2 looks like wrt serial
[16:01] <hrw> same as bb
[16:01] <ogra> which BB :P
[16:01] <rsalveti> c4
[16:01] <hrw> the only existing for customers one
[16:02] <ogra> XM just has a sub-d socket :)=
[16:02] <hrw> ogra: xM still is you-can-dream device rather
[16:02] <ogra> nah, they are on sale
[16:02] <hrw> finally?
[16:02] <hrw> good
[16:02] <ogra> big backlog, but they sell them
[16:02] <ogra> you should know !
[16:03]  * hrw waits for personal pandaboard anyway
[16:03] <ogra> i know some of your colleagues have A2's
[16:03] <hrw> ogra: infrastructure team has some toys, yes
[16:03] <hrw> ogra: I work on cross compiler packages so no linaro hw for me
[16:03] <ogra> and they were ordered in a std. way afaik
[16:04] <ogra> mean !
[16:04] <hrw> anyway before uds-n I should have panda @home for any experiments
[16:11] <jo-erlend> ogra, do you think I might have better results with another monitor?
[16:11] <ogra> i dont even know how well the kernel supports IEGP2 at all
[16:11] <ogra> especially the framebuffer code
[16:12] <ogra> we dont have such boards in the team so nobody tested it ... and the configuration tool on the image expects a display so you end up with a pretty much messed up image following the howto
[16:13] <rsalveti> I'd recommend you to try linaro's kernel
[16:13] <rsalveti> maybe you have a better result
[16:13] <ogra> yeah
[16:13] <ogra> though i think they are using them mainly with a minimal image for development
[16:13] <ogra> they might not care about graphics there
[16:13] <ogra> hey lag_
[16:14] <ogra> had a safe flight ?
[16:14] <rsalveti> but if it works, it should at least show something at his monitor
[16:14] <ogra> yeah
[16:14] <lag_> Hey orga
[16:14] <lag_> Yeah, 23hrs ...
[16:14] <rsalveti> lag_: out, taiwan?
[16:15] <rsalveti> *ouch
[16:15] <lag> Yeah
[16:15] <lag> :)
[16:15] <prpplague> rsalveti: looks like gnome-power-manager has an issue with my netbook
[16:16] <rsalveti> prpplague: hey, so did you try to disable it?
[16:16] <lag> Has anyone managed to get ftrace working on the Panda yet?
[16:16] <rsalveti> has anyone tried?
[16:16] <lag> I think so
[16:16] <lag> mpoirier was working on it last I heard
[16:17] <rsalveti> well, don't think mpoirier got his panda already
[16:18] <lag> I think he was working on omap generally omap/omap4
[16:18] <lag> If it works on OMAP it should work on OMAP4
[16:19] <rlameiro> lp #643626
[16:19] <ubot2> Launchpad bug 643626 in jackd-defaults (Ubuntu) "Jackd doesnt start on ARM Beagle board clone IGEPv2 (affects: 1) (heat: 6)" [Undecided,New] https://launchpad.net/bugs/643626
[16:20] <prpplague> rsalveti: yea, just turned it off so that it doesn't autoload on startup
[16:20] <prpplague> rsalveti: works fine after that
[16:20] <prpplague> rsalveti: i'll start debugging the issue this week
[16:25] <rlameiro> persia: lp #643626 , does it need more info ?
[16:25] <ubot2> Launchpad bug 643626 in jackd-defaults (Ubuntu) "Jackd doesnt start on ARM Beagle board clone IGEPv2 (affects: 1) (heat: 6)" [Undecided,New] https://launchpad.net/bugs/643626
[16:27] <rsalveti> prpplague: cool, please file a bug for it when you have time
[16:27] <rsalveti> just to let people know that we have issues with your hardware
[16:29] <hrw> ~curse omap3sd
[16:30] <rlameiro> mpoirier: Persia told me that you are into audio in ARM
[16:30] <mpoirier> rlameiro: yes, that is correct.
[16:30] <rsalveti> hrw: haha, giving apt-get update/upgrade is *painful* on on it
[16:30] <rlameiro> mpoirier: waht is your board?
[16:30] <hrw> rsalveti: 6-7h so far
[16:30] <rsalveti> ouch
[16:30] <mpoirier> I will get an 8 layer panda
[16:31] <rlameiro> mpoirier: lucky.... i have a IGEPv2
[16:31] <hrw> rsalveti: when it will finish this 4GB SD will be /boot only - rest on usb pata harddrive
[16:31] <rlameiro> having some problems with audio on it
[16:31] <mpoirier> rlameiro: did you get it from amitk ?
[16:32] <rlameiro> mpoirier: lp #643626  and #642465
[16:32] <ubot2> Launchpad bug 643626 in jackd-defaults (Ubuntu) "Jackd doesnt start on ARM Beagle board clone IGEPv2 (affects: 1) (heat: 6)" [Undecided,New] https://launchpad.net/bugs/643626
[16:32] <ubot2> Launchpad bug 642465 in puredata (Ubuntu) "Puredata outputs Very bad sound when using ALSA backend (affects: 1) (heat: 8)" [Undecided,New] https://launchpad.net/bugs/642465
[16:32] <mpoirier> rlameiro: again, did you get the board from amitk ?
[16:32] <rlameiro> mpoirier: amitk? I bought it from the builders - ISSE
[16:32] <mpoirier> rlameiro: ok.
[16:33] <rlameiro> what is amitk?
[16:33]  * amitk looks up
[16:33] <ogra> lol
[16:33] <rlameiro> ahhhh
[16:33] <rlameiro> lol
[16:33] <mpoirier> amitk works with me on the linaro side.
[16:33] <ogra> amitk, you are selling boards  ?!?
[16:33] <rlameiro> nice
[16:33] <amitk> ogra: had to supplement my income :-p
[16:34] <ogra> *giggle*
[16:34] <mpoirier> he and I had a run-in with and IGEPv2 board a couple of months ago.
[16:34] <rlameiro> amitk: website???
[16:34] <rsalveti> hrw: yep, at my devel board I'm using an usb disk
[16:34] <ogra> linaro.org
[16:34] <rlameiro> no, i tought amitk had a shop :D
[16:34] <rlameiro> lol
[16:34] <rlameiro> i know linaro
[16:34] <amitk> :)
[16:34] <mpoirier> rlameiro: i'm not surprised sound isn't great on IGEPv2.
[16:35] <rlameiro> mpoirier: well, the codec isn nothing special, but i cant use my edirol ua4fx  also
[16:35] <rlameiro> I dont know but it doesnt work
[16:36] <rlameiro> audio is the same that the beagle, the twl 4030
[16:36]  * rsalveti lunch
[16:36] <mpoirier> rlameiro: I know - it should work, until you find different setting in the board file or some place else...
[16:36] <rlameiro> i am thinking on making a custom audio board for it using the TI bus /12S
[16:37] <mpoirier> rlameiro: I currently have the same problem with a Gumstix overo.
[16:37] <mpoirier> rlameiro: related to display.
[16:39] <rlameiro> i never wanted to use the onboard audio codec , its too noisy, but if jack isnt starting i dont know if it will be of help, also plugin my edirol usb interface doesnt work either, it is detected but cant play anything, not even with aplay...
[16:40] <mpoirier> rlameiro: not cool.
[16:41] <rlameiro> mpoirier: i am feeling like a devil's advocate, 2 days runnig ubuntu on my board 2 bugs posted
[16:42] <rlameiro> and I didnt posted the edirol issue yet...
[16:42] <mpoirier> rlameiro: that is normal.
[16:42] <mpoirier> we haven't worked much with the IGEPv2 board.
[16:42] <rlameiro> well, but it seems to be a nice board
[16:43] <mpoirier> rlameiro: no doubt, no doubt friend.
[16:43] <rlameiro> 512/512 MB, 720 Mhz etc
[16:43] <mpoirier> rlameiro: i know.
[16:43] <rlameiro> wifi bluetooth and stuff
[16:44] <rlameiro> I prefer it way more to the beagle... and after buying it, the beagle XM came out....
[16:44] <rlameiro> but anyway since igepv2 come from spain, its cheaper for me here in Portugal
[16:45] <hrw> anyway it works
[16:45] <mpoirier> rlameiro: file the bugs you find - our agenda may permit to address them.
[16:45] <hrw> I remember playing with igep during platform sprint when we had one with SD not working under linux (working in uboot)
[16:45] <rlameiro> mpoirier: do you have igepv2 at canonical?
[16:46] <mpoirier> we have a few but again, ran in some SD card trouble with them.
[16:46] <mpoirier> rlameiro: back then I didn't have the time to address them.
[16:46] <rlameiro> ahh ok
[16:47] <mpoirier> Getting the boads aren't the problem - time to work on them is.
[16:47] <furibondox> hello
[16:47] <rlameiro> well have to go to work
[16:47] <rlameiro> cya latter everyone
[16:48] <furibondox> I've a problem with udev... it seems that udev doesn't read my custom rules
[16:48] <mpoirier> rlameiro: thanks for your comment and feedback on the IGEPv2.
[16:48] <furibondox> I'm using ubuntu lucid
[16:49] <furibondox> the rules are correct and worked fine in jaunty
[16:51] <furibondox> my concern is that during the init, the mount -o move /dev should be responsable of this behavior... once I am into the rootfs (after run-init) all devices are copied into the rootfs and the udev has already took effect but I'm not sure if it read the rules again or not (but seems not)
[16:53] <furibondox> again, if I manually run "udevadm control --reload-rules", my custom rules are not processed
[16:53] <furibondox> any idea?
[16:54] <ogra_ac> where do you put them ?
[16:54] <furibondox> in /etc/udev/rules.d/
[16:55] <ogra_ac> that should work
[16:55] <furibondox> ogra_ac: just another question... if you want I can send you my rootstock script with 2 little mods
[16:56] <ogra_ac> file a bug, attach the patch
[16:56] <ogra_ac> i'm not doing much on rootstock atm
[16:56] <furibondox> I've never submit a bug for ubuntu, where should I do it?
[16:57] <ogra_ac> so are you using update-initramfs now  ?
[16:57] <ogra_ac> (for bugs see /topic ;) )
[16:57] <furibondox> well... now I'm using the same initrd as the ubuntu but w/o update-initramfs
[16:58] <ogra_ac> i'm not sure, but it might copy the rules into teh initramfs
[16:58] <furibondox> I simply copied it from the one present after the rootstock (used with the beagle kernel) and then I modified it a little to fit our needings
[16:58] <furibondox> but is almost the same
[16:59] <furibondox> except little hw configuration
[16:59] <ogra_ac> did you compare with an unmodified one ?
[16:59] <furibondox> yes...
[16:59] <furibondox> is the almost the same
[16:59] <ogra_ac> and that doesnt read your rules either ?
[17:00] <furibondox> I've not the original initrd now so I can't try to do it
[17:00] <furibondox> I will try to set the rules in the initrd
[17:00] <furibondox> tnx
[17:00] <furibondox> now I must go...
[17:00] <furibondox> see you tomorrow
[17:01] <furibondox> bye
[17:01] <ogra_ac> ciao
[17:01] <furibondox> ciao ;)
[18:58] <ogra_ac> oh my
[18:59] <ogra_ac> the android browser refuses to open https links with selfsigned certs
[19:01] <GrueMaster> Yes, this is known.
[19:01] <armin76> feature
[19:01] <GrueMaster> ogra_ac: Use opra for android.
[19:01] <ogra_ac> thats unusable
[19:02] <ogra_ac> i have it installed
[19:07] <ogra_ac> opera is just painful to read
[19:08] <ogra_ac> it renders awfully
[19:08]  * ogra_ac wishes there ws afirefox build
[19:08] <ogra_ac> *was
[19:10] <ogra_ac> or chromium
[19:29] <GrueMaster> Hey, hostname has returned to oem-config.  Nice.
[20:29] <slangasek> GrueMaster: u-boot-linaro made it over the weekend - how does it look on omap4?
[20:29] <rsalveti> slangasek: working fine at our current daily image
[20:29] <rsalveti> leds are now working
[20:30] <rsalveti> no other issues until now
[20:30] <GrueMaster> I booted today's image.  Serial console shows U-boot 2010.09-rc1 (Sep 18 2010 - 11:15:35) for u-boot.
[20:30] <GrueMaster> Looks good.  And I have blinking lights next to the SD slot.  :D
[20:31] <ogra_ac> is there a chance that we can make it blink a little slower ?
[20:31] <slangasek> rsalveti, GrueMaster: awesome to hear, thanks
[20:32] <rsalveti> ogra_ac: heartbeat? need to check the code
[20:32]  * ogra_ac would prefer once a second
[20:32] <rsalveti> depends on the cpu load currently
[20:32] <ogra_ac> ah
[20:32] <ogra_ac> higher load means faster ?
[20:33] <rsalveti> yup
[20:33] <rsalveti> that's why heartbeat
[20:33] <ogra_ac> then we need to improve the userspace !
[20:36] <rsalveti> ogra_ac: playing with android?
[20:36] <rsalveti> or using ubuntu already?
[20:37] <GrueMaster> So, is the heartbeat tied to actual cpu load or all load?  Don't need excessive flutter waiting on SD I/O.
[20:37] <rsalveti> don't remember the code, but it should be cpu load
[20:38] <ogra_ac> rsalveti, i bricked the device on sun. so i'm a bit more careful now :)
[20:38] <rsalveti> hahah, got it :-)
[20:38] <ogra_ac> (with teh replacement)
[20:39] <ogra_ac> luckily the guys at the discounter where they sell it are totally clueless so it was easy to get a replacement
[20:39] <rsalveti> hehe, cool
[20:50] <GrueMaster> slangasek: Was that u-boot omap4 only?
[20:57] <ogra_> grmblfjx
[21:09] <GrueMaster> Yea.  My XM is finally working again (so far).  3rd time trying a fresh daily image today.  Currently finishing oem-config stage.
[21:09] <devilhorns> woot ! :)
[21:11] <GrueMaster> And...it crashed.
[21:11] <GrueMaster> sigh.
[21:11] <devilhorns> must be because I said something, sorry ;)
[21:12] <GrueMaster> It's possessed.
[21:12] <devilhorns> hehe
[21:13] <ogra_ac> kill the daemons
[21:13] <slangasek> GrueMaster: there's an omap3 as well, but you guys aren't using that yet
[21:14] <ogra_ac> slangasek, yeah, i havent switched over yet
[21:14] <slangasek> GrueMaster: you're welcome to test it, however: u-boot-linaro-omap3-beagle
[21:14] <ogra_ac> will do so next time i touch debian-cd
[21:14] <devilhorns> ogra_ac, real quick ... any hardware news for me ?
[21:14] <ogra_ac> devilhorns, ask davidm ?
[21:14] <ogra_ac> he has it
[21:15] <GrueMaster> slangasek: Ok, now I know why XM was reporting a different version.  Thanks.
[21:15] <devilhorns> ogra_ac, haven't seen him on yet today, but I'll ask when I see him, thanks :)
[21:15] <ogra_ac> Gruemaster; which Xm version did you have ?
[21:15] <GrueMaster> P8
[21:16] <ogra_ac> you might need to boot with mem="%&M
[21:16] <ogra_ac> grr
[21:16] <ogra_ac> 256M
[21:16] <GrueMaster> Extra flaky goodness.
[21:16] <ogra_ac> i'd love to know why the kbd always does that
[21:16] <GrueMaster> Actually, it was running fine before Friday's power glitch.
[21:17] <ogra_ac> ouch
[21:17] <GrueMaster> Oddly enough, it is behind two surge protected power strips.
[21:17] <ogra_ac> well
[21:18] <ogra_ac> wait for the a2
[21:18] <GrueMaster> I spent all day Friday trying to get Beta to reinstall.
[21:18] <ogra_ac> p8 is really not worth to invest much time into it
[21:18] <GrueMaster> I may have hosed my microSD cards or something.
[21:18] <ogra_ac> i know david is on it
[21:18] <GrueMaster> I do what I can with what I have.
[21:19] <GrueMaster> Yea, he mentioned it this morning.
[21:19] <ogra_ac> sure but the hw is known to be flaky
[21:20] <rsalveti> yep, you shouldn't waste too much time on your xM
[21:20] <rsalveti> I always get some weird results with mine
[21:20] <rsalveti> even with mem=265M
[21:20] <rsalveti> 256
[21:21] <GrueMaster> As I said, I haven't had issues until Friday.
[21:22] <ogra_> grmpf
[23:26] <GrueMaster> Well, it is now official.  I can not get audio out of the default apps on any of our armel platforms in Maverick.
[23:26] <GrueMaster> I can at least get speaker-test to work on beagle with some massaging in alsamixer, but no gui apps.
[23:26] <GrueMaster> And now dove is down.
[23:27] <GrueMaster> And not a peep from my panda ES2 8L.
[23:27] <dcordes> GrueMaster: No good :( Unfortunately I don't have the working alsa yet else I could test
[23:27] <dcordes> I'm running 19th preinstalled
[23:28] <GrueMaster> I'm running today's (20100920).  Devices are visible, but nothing works.
[23:29] <dcordes> Maybe some pulseaudio thing ?
[23:29] <GrueMaster> No.  speaker-test can bypass it and go directly to hw.
[23:30] <dcordes> Well if that is the only thing that works ...
[23:30]  * dcordes shrugs
[23:31] <dcordes> How come all the gnome-panel applets are fixed now and what is the correct way to change this ?
[23:31] <dcordes> Couldn't find settings in  ~
[23:31] <persia> Could you give more context?
[23:32] <persia> GrueMaster, What did you need to do in alsamixer to get the beagle to work?  Is this different from what you had to do before?
[23:34] <GrueMaster> On beagle, in alsamixer, I had to unmute DAC2 Analog and crank it to 100, Crank Headset to 100, and unmute HeadsetL Mixer Audio L2 & HeadsetR Mixer Audio R2.
[23:35] <GrueMaster> That will at least get speaker-test to work (that's the cli speaker-test with -c 2 -t wav options).
[23:36] <persia> That's enough :)
[23:36] <persia> mpoirier, Did you have a chance to look at the beagleboard kernel-side config?  Which amps are turned on there?  Which have sensible volumes?
[23:37] <mpoirier> persia: not haven't yet - I'm currently setting up my board.
[23:37] <mpoirier> persia: I got sound going though using the scripts.
[23:37] <persia> Oh.  My apologies.  I thought you'd had a working beagle for some time.
[23:38] <mpoirier> persia: i have a working beagle indeed.  Just received my panda today.
[23:38] <mpoirier> persia: I'm on panda now.
[23:39] <persia> Ah, and you don't easily have both wired to audio.  I can understand that :)  Anyway, figured that with GrueMaster having just gotten audio to work post-fiddling would be a handy time to get it to work pre-fiddling kernel-side.
[23:42] <persia> GrueMaster, Have you tried USB audio also?  Are you able to reproduce rlamiero's report?
[23:42] <mpoirier> persia: I'll address panda first.
[23:42] <GrueMaster> I don't have any usb-audio devices.
[23:42] <persia> mpoirier, OK.
[23:42] <persia> GrueMaster, Aha.  Then no, and because you can't.  Thanks for trying :)
[23:43] <GrueMaster> Oh, wait.  I bought a converter several years ago for hooking up to a record player.  Let me find it.
[23:43] <mpoirier> persia: where's the u-boot git tree we are using for the omap4 image ?
[23:43]  * persia has no idea
[23:43] <persia> I *think* we're using the linaro uboot package.
[23:44] <GrueMaster> mpoirier: it is the linaro u-boot.  slangasek would know.
[23:44] <persia> I don't know where the upstream git tree lives.  Note that there is at least one Ubuntu-specific patch against upstream applied, so you may want to examine the source package.
[23:45] <jo-erlend> is there any difference between distros in hardware support for omap3 devices like IGEPv2?
[23:46] <mpoirier> jo-erlend: we haven't had a chance to really fit ubuntu to the IGEPv2 board.
[23:47] <jo-erlend> so I've been told.
[23:47] <mpoirier> jo-erlend: there's always stuff in the board specific file that needs to be adjusted.
[23:48] <jo-erlend> board spesific file?
[23:48] <jo-erlend> I'm new to these ARM-stuff. :)
[23:48] <mpoirier> jo-erlend: each board is different - same processor but different mapping of devices.
[23:48] <persia> jo-erlend, the kernels differ between distributions, meaning that the specific handling of various devices tends to differ slightly.  Everyone tries to feed upstream, but it may take more or less time in some places or other.
[23:49] <persia> Userspace should have nearly precisely the same level of hardware support everywhere.
[23:49] <jo-erlend> ok, so I might be able to get it working by simply replacing the kernel with one from another distro?
[23:49] <persia> If you're a kernel person, please use Ubuntu, hack the kernel to work for you, and report how the kernel needs changing.
[23:49] <persia> Potentially, although we (obviously) can't support someone else's kernel.
[23:50] <persia> Note that there are at least 2 omap kernels in Ubuntu, so you get to try both before you go somewhere else :)
[23:51] <persia> If you aren't a kernel person, you ought be able to use Ubuntu userspace with some other kernels, but you do need to have some specific configuration options set: you may end up recompiling if the folks who produce your kernel don't set those.
[23:51]  * ogra_ac would recommend trying the linaro kernel
[23:51] <jo-erlend> Thanks. There is a wiki page about running ubuntu on the igep. This tells me that someone there must have been successful. That's encouraging, I think.
[23:51] <persia> Why would you expect it to be better than the mainline kernel?
[23:51] <slangasek> mpoirier, GrueMaster: git://git.linaro.org/boot/u-boot-linaro-stable.git
[23:52] <ogra_ac> because linaro uses iegp2 for development
[23:52] <mpoirier> fantastic slangasek - thanks.
[23:52] <ogra_ac> while we dont have a single of these boards
[23:57] <dcordes> is ac nvidia smartbook ?