[00:02]  * xnox should not have started reinstalling
=== smplman_ is now known as smplman
=== heathkid|2 is now known as heathkid
=== heathkid|2 is now known as heathkid
=== doko_ is now known as doko
[09:43] <nigelr> ]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]
[09:43] <nigelr> sorry
[09:43] <nigelr> screen doing some weird stuff there.
[09:53] <ppisati> ogra_: todays omap4 live are still 3.4 based, how do i check if they promoted to 3.5?
[09:53] <ogra_> ppisati, look at the .manifest fiel for the image
[09:54] <ogra_> *file
[09:54] <ppisati> ogra_: you call it "seed" right? the thing that list all the packages that should be included in an image?
[09:54] <ogra_> or for desktop it might be .list
[09:54] <ogra_> the seed is the input for tasks and metapackages, yeah, it defines what goes where
[09:55] <ogra_> but the .list and .manifest files show you what actually ended up in the image ;)
[09:55] <ppisati> well
[09:55] <ppisati> http://cdimage.ubuntu.com/daily-live/current/quantal-desktop-armhf+omap4.manifest
[09:55] <ppisati> shows me 3.4 since the live is 3.4 based
[09:55] <ogra_> linux-image-3.4.0-204-omap4	3.4.0-204.9
[09:56] <ogra_> The following packages have unmet dependencies:
[09:56] <ogra_>  empathy : Depends: empathy-common (= 3.5.5-0ubuntu1) but 3.5.90-0ubuntu2 is to be installed
[09:56] <ogra_> well, no image builds since two days
[09:56] <ogra_> yesterdays failed on pythin-glib i think
[09:56] <ppisati> for emapthy?
[09:56] <ppisati> :(
[09:56] <ogra_> todays on emptahy
[09:56] <ogra_> try the server image ;)
[09:56] <ppisati> where do you check these status?
[09:57]  * ppisati is missing some ml... 
[09:57] <ogra_> nah, they are hidden on people.ubuntu.com
[09:57] <ogra_> i'm in the cdimage team, we get the logs mailed daily
[09:57] <ogra_> one sec
[09:58] <ogra_> http://people.canonical.com/~ubuntu-archive/cd-build-logs/
[09:59] <ogra_> and http://people.canonical.com/~ubuntu-archive/livefs-build-logs
[09:59] <ppisati> cool
[09:59] <ogra_> todays is at http://people.canonical.com/~ubuntu-archive/livefs-build-logs/quantal/ubuntu-omap4/20120822/livecd-20120822-armhf.out
[09:59] <ppisati> the last one is the one i was looking for
[10:02] <ppisati> ogra_: when does it run next?
[10:04] <ogra_> ppisati, hmm, good question, the crontab on the build machine looks a bit weird
=== Quintasan_ is now known as Quintasan
[13:59] <ppisati> The following packages have unmet dependencies:
[13:59] <ppisati>  gwibber : Depends: gwibber-service (= 3.5.2-0ubuntu1) but 3.5.3-0ubuntu1 is to be installed
[13:59] <ppisati>  gwibber-service : Conflicts: libgwibber2 but 3.5.2-0ubuntu1 is to be installed
[13:59] <ppisati>                    Breaks: libgwibber2 but 3.5.2-0ubuntu1 is to be installed
[13:59] <ogra_> fun
[13:59] <ppisati> ogra_: ^
[14:00] <ogra_> say thanks to people that upload packagesets to the main archive instead of proposed
[14:00] <ogra_> it should resolve itself after a while
[14:00] <ppisati> yes, but i can't test my kernel in the archive
[14:01] <ogra_> our kernel uses gwibber ?
[14:01] <ppisati> no
[14:02] <ogra_> (does it twitter dmesg lines ?)
[14:02] <ppisati> but builders don't complete a build if $foobar is brokern
[14:02] <ogra_> so just grab a workin server image then
[14:02] <ppisati> i want the desktop image
[14:02] <ogra_> and apt-get install the kernel
[14:02] <ogra_> forget desktop
[14:02] <ogra_> unusable atm
[14:02] <ogra_> even if it would build
[14:03] <ppisati> ah right
[14:03] <ppisati> bah
[14:03]  * ppisati dies in a corner
[14:03]  * ogra_ does that since a week
[14:03] <ogra_> welcome to the club
[14:08] <ogra_> gwibber (3.5.3-0ubuntu2) quantal; urgency=low
[14:08] <ogra_>   * debian/control:
[14:08] <ogra_>     - add some missing internal package dependencies
[14:08] <ogra_> ppisati, ^^^
[14:08] <ogra_> uploaded 8 min ago
=== jibel__ is now known as jibel
[14:59] <rsalveti> ogra_: I'm now working on the remaining changes for pvr, should have the packages ready later today
[15:01] <ogra_> \o/
[16:18] <GrueMaster> What?  Arm daily images are broken?  Say it isn't so.
[16:19] <ogra_> GrueMaster, nah, it isnt ... server always works :P
[16:19] <GrueMaster> Pfft.  Of course it works, there's practically nothing there.
[16:22] <ppisati> besides, where're the server images now?
[16:22] <ppisati> https://wiki.ubuntu.com/ARM/Server/Install
[16:22] <ogra_> where they always were
[16:22] <ppisati> refers to P
[16:23] <ogra_> cdimage.ubuntu.com/ubuntu/server/daily/
[16:23] <ppisati> ah
[16:23] <ogra_> err
[16:23] <ogra_> ubuntu-server
[16:23] <ppisati> Not Found
[16:23] <ogra_> (sorry trying to get used to a new kbd here)
[16:23] <ppisati> 404
[16:23] <ogra_> http://cdimage.ubuntu.com/ubuntu-server/daily/
[16:23] <ppisati> no ok
[16:24] <ogra_> my daily omap3 autotest just finished ...
[16:24] <ogra_> i would expect omap4 to be fine as well ... unless someone uploaded a kernel that doesnt work on servers :P
[16:25] <ogra_> ppisati, dont forget to mount the first partition after dd'ing and copy preEnv.txt-serial over preEnv.txt
[16:25] <ogra_> unless you want a framebuffer based install
[16:26] <ppisati> if (am_i_on_a_server()) panic("thou shalt not server\n");
[16:26] <ogra_> haha
[16:26] <ogra_> dont let Daviey see that :)
[16:26] <ppisati> ack
[16:28] <GrueMaster> I wonder if u-boot could be made to detect if there is a monitor attached?  Something like "if EDID {console=fb} else {console=serial}" (simple logic).
[16:28] <ogra_> not in hush shell script
[16:28] <GrueMaster> Not that I do much with my pandas these days (except intercept calls from ET).
[16:28] <ogra_> it already can drive the framebuffer
[16:29] <ogra_> (see the beagle images that get this over-guly orange screen before loading the kernel)
[16:29] <ogra_> the prob is that your u-boot.bin gets so big
[16:29] <GrueMaster> I know it can drive the framebuffer, but it would be nice if it detected something on the other end.
[16:30] <GrueMaster> So?  It is on SD.
[16:30] <ogra_> to be loaded into ram
[16:30] <ogra_> MLO needs to assign more space in that case etc etc
[16:31] <GrueMaster> Doesn't it get booted out once the kernel boots?
[16:31] <ogra_> yep
[16:31] <infinity> Yeah, it's a question of how much RAM you have to play with before the kernel does anything to it.
[16:31] <infinity> And mapping it sanely.
[16:31] <GrueMaster> I'm not saying for u-boot to actually drive the FB, just detect if it is there.  Let the hush shell do simple scripting (which I know it already can do most of what I recommended).
[16:31] <infinity> FSVO "sane".
[16:32] <infinity> Which basically is just rough arithmetic and some luck.
[16:32] <infinity> And I'm pretty sure you'd need the FB driver built-in to even hope to detect a display on the other end.
[16:32] <infinity> But maybe not.
[16:32] <ogra_> GrueMaster, its not the hush shell, you need to include the whole framebuffer stack to even provuide that device somewhere
[16:32] <infinity> Seems like a bit of effort for just one subarch, though.
[16:33] <infinity> (I agree that it would be nice to autodetect the console, mind you, I've brought it up before too)
[16:33]  * ogra_ grumbles about that new kbd
[16:33] <GrueMaster> Do you need the FB stack to do plug detection?
[16:33] <ogra_> GrueMaster, yes
[16:34] <ogra_> at least parts of it, you can probably cut it down if you dont want to display stuff
[16:34] <GrueMaster> Hmm.  I would think a snippet of code in u-boot would be enough.
[16:34] <ogra_> but thats really a lot effort for not much benefit
[16:34]  * ogra_ thinks since we dropped boot.scr it got easy enough to edit 
[16:35] <ogra_> i'll probably drop the preEnv.txt-serial too and just put the info on the wiki
[16:36] <ogra_> (its not like its actually hard to add "console=ttyO2" to the end of a file that only contains one line
[16:36] <ogra_> )
[16:48] <xnox> ogra_: you'd be surprised =)
[16:49] <ogra_> heh
[16:49] <xnox> ogra_: help.ubuntu.com & wiki.ubuntu.com & askubuntu.com are full of duplicate one-liners....
[16:49] <ogra_> hahaha, complain to the doc team :)
[16:50]  * ogra_ didnt even know there are people asking arm questions on askubuntu
[16:52] <ogra_> (and i wonder who answers them ?)
[16:54]  * GrueMaster really wishes Unity would support rdp.
[16:55]  * ogra_ really wishes unity would work on arm :P
[16:57] <GrueMaster> Yea, that too.
[17:02] <rsalveti> I don't think console autodetection should happen based on the monitor presence
[17:02] <rsalveti> that's something that can be done, but then you'd need to boot the board with the monitor already plugged
[17:03] <rsalveti> and even at my own use cases, that doesn't happen so frequently
[17:03] <infinity> rsalveti: Based on monitor and keyboard presense is a clever way to do it for installers.  Not for installed systems.
[17:03] <prpplague> anyone making it out to LPC/LCNA ?
[17:03] <infinity> rsalveti: That's the trick generally pulled on other platforms.  Autodetect for the installer, trust the bootloader's set default after that.
[17:03] <rsalveti> I'll be at plumbers
[17:03] <infinity> prpplague: I'm plumbing.
[17:04] <prpplague> infinity: ahh dandy
[17:04]  * ogra_ would like to, but forgot to ask for compan coverage
[17:04] <prpplague> infinity: come on by the TI booth if/when you have time
[17:04] <ogra_> *company
[17:04] <infinity> prpplague: Do you have presents?
[17:04] <prpplague> infinity: just one at linuxcon
[17:04] <ogra_> mouse mats with pandas on them :)
[17:04] <rsalveti> infinity: I understand the use case, but I'm still not convinced that mouse+monitor should be a reference for automatically selecting the console here
[17:04] <infinity> (Is LC co-located with LPC, I haven't really looked at the former)
[17:05] <rsalveti> it'll depend a lot based on each soc again
[17:05] <rsalveti> which can be a huge pain
[17:05] <rsalveti> can't be something that the kernel/initrd would be able to detect?
[17:05] <infinity> rsalveti: Eh.  It's as correct as any other guess for installer first-boot.
[17:05] <rsalveti> once we're at the kernel side, it's a lot easier
[17:06] <prpplague> infinity: http://lcna2012.sched.org/event/df97704300418c3eee409c7254c5160d#.UDURbWY6css
[17:06] <rsalveti> prpplague: will you be at plumbers as well?
[17:06] <rsalveti> oh, cool
[17:06] <prpplague> rsalveti: yea
[17:07] <prpplague> rsalveti: no presentation though
[17:07] <ogra_> present yourself :)
[17:07] <prpplague> hehe
[17:07] <infinity> prpplague: Oh, you misunderstood, I asked if you had *presents*, not a presentation. ;)
[17:07] <prpplague> infinity: ahh
[17:07] <infinity> prpplague: As in, gifts, toys, incentive to come say hi. ;)
[17:07] <rsalveti> :-)
[17:07] <prpplague> infinity: nothing special at linuxcon/lpc
[17:07] <GrueMaster> I'd go out for post-event beers, but it is in San Diego, not Portland.
[17:08] <prpplague> infinity: i have some plans for some stuff at ELCE
[17:08] <infinity> prpplague: Oh well.  Maybe I'll find the time to come be friendly anyway. :P
[17:08] <ppisati> who is going to ELC in Barcelona?
[17:08] <prpplague> infinity: i can't afford to give a bunch of stuff away at every conference, cuts in to my beer money
[17:08]  * ppisati is going
[17:08] <infinity> rsalveti: Bring a box of Linaro pens, we can take turns throwing them at prpplague.
[17:08] <ogra_> ppisati, its such a bad timin :/
[17:08] <ogra_> *timing
[17:09] <ppisati> ogra_: why? right after uds
[17:09] <ogra_> exactly :)
[17:09] <rsalveti> infinity: yeah, I think I have a bunch around
[17:09] <infinity> Right after UDS is perfect conference timing, no one works that week anyway.
[17:09] <ppisati> come on, you can drive down from Germany :)
[17:09]  * ogra_ will go by car to UDS ... 
[17:09] <ogra_> ppisati, no, from denmark
[17:09] <prpplague> yea LF has done a terrible job of scheduling conferences this year
[17:10] <GrueMaster> Bloody hell.  Remote desktop IS available, just not enabled.  Grrr.  http://liberiangeek.net/2012/04/list-of-remote-access-support-software-for-ubuntu-12-04-precise-pangolin/ (see bottom).
[17:10] <ppisati> unfortunately i'll miss plumbers this year
[17:10] <ppisati> but at least i'll go to elc
[17:11]  * ogra_ curses bluetooth 
[17:11] <ogra_> damned
[17:13]  * GrueMaster now wishes his laptop would support more than 2 monitors.  So many remote machines to work with, so little desktop space.  sigh.
[17:13] <prpplague> ugh, i so hate the lack of output on ubuntu desktop installs
[17:14] <ogra_> prpplague, well, we improved that, now we dont even start any GUI in quantal anymore
[17:16] <prpplague> ogra_: 11.10 works with pandaboard-es and dvi right?
[17:16] <ppisati> ogra_: so i din't get, are you coming or not to elc? and if yes, when? friday? sat?
[17:16] <ogra_> ppisati, no, i wont
[17:17] <infinity> prpplague: I don't think we supported the ES until 12.04?  I could be wrong.
[17:17] <ogra_> unless someone thinnks they need to send me
[17:17] <ogra_> we supported the ES with lucid iirc
[17:17]  * prpplague hates when management asks for ubuntu installations for demos
[17:17] <GrueMaster> infinity: It was supported (barely) in 11.10
[17:18] <GrueMaster> Lucid never supported omap.
[17:18] <ogra_> oh, it did, not omap4 though
[17:19] <GrueMaster> omap3 support was a tech preview iirc.  And very late in the game.
[17:19] <infinity> GrueMaster: Yeah, I know for a fact that natty doesn't even pretend to boot on an ES, never tried oneiric.
[17:20] <ogra_> no, right, it was maveric
[17:20] <ogra_> that was the release we spent at TI during release week
[17:20] <ogra_> to bring up the ES iirc
[17:20] <GrueMaster> As far as DVI goes, I'm not sure if it was supported properly (or is now).  afaik, the images assume output on hdmi port only, and there was no dual monitor output.
[17:21] <GrueMaster> ogra_: That was the original panda omap4 4450.  I thought the ES was the 4460.
[17:21] <scientes> GrueMaster, hdmi is backwards compatible with DVI-D
[17:21] <GrueMaster> (or do I have those numbers wrong).
[17:22] <GrueMaster> scientes: I know, but there are two "hdmi" ports on panda, 1 is true HDMI, the other is DVI-D in an hdmi form factor.
[17:22]  * prpplague wishes he had time to fix these issues
[17:22] <scientes> hdmi also has usb and ethernet for the crazies
[17:23] <GrueMaster> Only in newer HDMI specs.
[17:23] <scientes> and for the crazies
[17:23] <ogra_> GrueMaster, ES was the second gen panda
[17:23] <ogra_> iirc
[17:23] <infinity> ogra_: The ES definitely won't work on maverick or natty.
[17:23] <ogra_> hmm
[17:23] <infinity> ogra_: And the ES is the 4460, the other Pandas are 4430.
[17:23] <ogra_> what am i getting wrong here
[17:23] <GrueMaster> ogra_: Ok, then that wasn't supported until Oneiric
[17:24] <ogra_> oh, i mixed up E and ES
[17:24] <ogra_> and EA
[17:24] <GrueMaster> Yea, the EA was the preproduction 4430.
[17:24] <ogra_> EA was maverick til natty ... E natty and later
[17:24] <GrueMaster> The ES is the 4460.
[17:24] <ogra_> right
[17:24] <ogra_> sorry, my fault
[17:25] <GrueMaster> How is it that I'm the only one that always remembers this?  :P
[17:26] <ogra_> dunno, beer imprints in your braincells we others miss ?
[17:26] <GrueMaster> Heh.  That must be it.
[17:28] <rsalveti> :-)
[17:29] <rsalveti> E1, EA1, 6 layers, 8 layers ES2, lots of names for the older panda :-)
[17:29] <rsalveti> but yeah, 4460 is supported at precise already, and probably just with hdmi
[17:29] <rsalveti> dvi is only stable enough to be used since the 3.4 based kernel
[17:30] <rsalveti> still without support for dual monitors, something maybe prpplague would know more
[17:30] <rsalveti> but agreen said it'd be quite hard to sync both outputs to make them usable at the same time
[17:31] <prpplague> rsalveti: it isn't that hard to sync up, but does require some work
[17:31] <prpplague> rsalveti: the main problem is most of the things i need to test are via the DPI , not HDMI
[17:31] <prpplague> rsalveti: which makes using the builds almost useless
[17:31] <rsalveti> yeah, ok
[17:32] <prpplague> just very frustrating that my angstrom builds work great, but management only wants to see ubuntu
[17:35] <rsalveti> prpplague: and what is your issue with ubuntu at this point?
[17:36] <rsalveti> what is causing you headaches?
[17:36] <prpplague> rsalveti: no DPI support, basically the same issue i always have with it
[17:37] <rsalveti> I think that's supported with the latest kernel I'd guess
[17:37] <rsalveti> you could also use the ubuntu-linaro pre-built rootfs we have based on precise
[17:37] <rsalveti> which delivers the ti lt kernel
=== dduffey_afk is now known as dduffey
=== ogra__ is now known as ogra_
[17:38] <rsalveti> it's just that we had issues with it for a few kernel releases
[17:40] <rsalveti> if latest 3.5 kernel doesn't support it, it's probably config bug or similar
[17:42] <balloons> ogra_, does opengl|es mean unity for us on pandaboards? http://smspillaz.wordpress.com/2012/08/22/delivering-compiz-and-unity-on-the-next-wave-of-embedded-form-factors/
[17:42] <balloons> or do we still need the powervr driver?
[17:42] <prpplague> rsalveti: got a link to the linaro-ubuntu you are refering to?
[17:43] <rsalveti> balloons: we still need the pvr driver, but that's also landing today/tomorrow
[17:43] <ogra_> balloons, well, unity on x86 supports GL ... would it run without a GL capable xserver ? :)
[17:43] <rsalveti> which will allow us running unity 3d on panda
[17:43] <rsalveti> just like we had with precise, but now without a huge package patch :-)
[17:44] <rsalveti> prpplague: http://snapshots.linaro.org/precise/pre-built/lt-panda-x11-base/260/lt-panda-x11-base-precise_ubuntu-desktop_20120821-260.html
[17:44] <rsalveti> prpplague: download, dd it and let me know if it worked or not for you
[17:45] <balloons> ogra_, rsalveti thank you.. I assumed we still needed the driver. but that's awesome it's showing up soon as well
[17:45] <balloons> so friday's builds, fingers crossed should have all the goodness in them?
[17:45] <ogra_> yeah, by end of the week we should have all bits we need in the archive
[17:45] <balloons> wah-hoo!
[17:45] <jimerickson> yay!
[17:46] <ogra_> but they will need additional integration work (teh driver isnt seeded yet etc)
[17:46] <ogra_> so i'd rather say monday
[18:06] <prpplague> rsalveti: thanks will do
=== zyga is now known as zyga-afk
[18:41] <prpplague> rsalveti: looks like i am going to have to build a custom kernel to do what i need
[18:41] <prpplague> rsalveti: oh well
[18:52] <prpplague> rsalveti: do you know where the kernel source is for that linaro build?
[18:55] <prpplague> rsalveti: nm, looks like the linaro build is going to work, still testing
[19:04] <prpplague> rsalveti: doh detecting the wrong resolution, hehe
[19:45] <samek> is there any repository with gcc 4.4.4 available for ubuntu 10.04?
=== zyga-afk is now known as zyga
[20:34] <rsalveti> prpplague: but did it work at some leve at least?
[20:35] <prpplague> rsalveti: yea the linaro-ubuntu boots, but with the wrong resolution detected
=== jibel_ is now known as jibel
[23:37] <samek> i'm having trouble configuring serial console login on my arm board
[23:37] <samek> https://help.ubuntu.com/community/SerialConsoleHowto
[23:37] <samek> this doesn't work
[23:37] <samek> is there any updates to this document
[23:46] <samek> oh it works.. was using wrong tty device
[23:46] <samek> thanks for not answering ;)
[23:56] <GrueMaster> You're welcome.  :P