[08:21] <linu2> hi i have used kernel-3.10,and enabled framebuffer but when i try to run a sample qt application in at91sam9x5ek board using same kernel that says http://pastebin.com/YZxgptL4   can you tell mw what is that issues
[10:00] <aditya1> Hello all... i understand i am asking a repeated question... but i am looking out for a easier solution. 	i followed https://wiki.ubuntu.com/Nexus7/Installation to get my nexus running with ubuntu. The ubuntu installer looks for a different URL to download packages
[10:01] <aditya1> and manual steps are going for a hang... nothing is getting copied... is there a way to get ubuntu nexus installer installing from a local bootimg and .img files?
[10:19] <wookey> infinity: (or anyone): I made an image from the saucy arm64 tarball at http://cdimage.ubuntu.com/ubuntu-core/releases/13.10/release/ubuntu-core-13.10-core-arm64.tar.gz and booted it with my existing-working kernels. It does a load of upstart stuff and hangs at:
[10:19] <wookey> [   14.415224] init: startpar-bridge (procps-virtual-filesystems-started) goal changed from stop to start
[10:19] <wookey> anyone got any idea what that might mean?
[10:29] <hrw> nope
[10:36] <linu> hi i have used kernel-3.10,and enabled framebuffer but when i try to run a sample qt application in at91sam9x5ek board using same kernel that says http://pastebin.com/YZxgptL4   can you tell mw what is that issues
[10:46] <wookey> hrw: do you have the magic runes you made that linux-boot .axf from?
[10:46] <wookey> I only ever used yours so have no idea how to maker a new one
[10:47] <linu> hi any idea about my issues
[10:47] <wookey> I just tried booting with img-foundation from http://releases.linaro.org/13.09/openembedded/aarch64/ and that doesn;t get far at all and backtraces
[10:47] <hrw> wookey: you mean Image -> axf ones?
[10:47] <wookey> yes
[10:48] <hrw> moment
[10:48] <wookey> I have a checkout of boot-wrapper-aarch64 which may be important
[10:50] <hrw> wookey: http://git.linaro.org/gitweb?p=openembedded/meta-linaro.git;a=blob;f=meta-linaro/recipes-kernel/linux/linux-linaro-fastmodel.inc;h=284f912ca5d5a9b29345eb6786fadfca837438b6;hb=HEAD
[10:50] <hrw> wookey: check do_compile_append()
[10:50] <hrw> I want aarch64 hardware...
[10:50] <wookey> cheers.
[10:51] <wookey> yes. me too. some lucky people have some now...
[10:51] <wookey> this is how youi know if you are important or not :-)
[10:55] <hrw> if you say so ;D
[10:55] <hrw> wookey: check demo tuesday at linaro connect
[10:59]  * XorA has not aarch64 goodness :-(
[10:59] <wookey> the schedule has teu afternoon mostly empty
[10:59] <wookey> tue
[14:30] <wookey> doko: riku: OK. I can boot the image if I don't run upstart
[14:32] <suihkulokki> wookey: perhaps some kernel option that is needed by upstart is not enabled in linaro kernel?
[14:32] <wookey> maybe, except that it worked fine with raring image
[14:32] <wookey> so what's changed? I guess the default kernel has, so maybe some assumptions
[14:33] <wookey> this one is 3.8.0
[14:33] <wookey> using the current linaro OE release img.axf barfs quickly with a backtrace
[14:34] <wookey> I should probably try latest model too
[14:35] <suihkulokki> wookey: note that img.axf is for fast model while img-foundation.axf is for foundation model
[14:39] <wookey> OK. I wondered what the diff was
[14:39] <wookey> I think I tried the latter - lets check
[14:41] <wookey> OK. prom,iosing - that just wants: Please append a correct "root=" boot option
[14:41] <wookey> how do I edit the command line in a .axf?
[14:41] <suihkulokki> you don't
[14:41] <suihkulokki> you rebuild your axf
[14:42] <wookey> OK. that's what I feared
[14:42] <suihkulokki> incidentally, different command line is the only difference between those to axf files
[14:42] <wookey> right. I had a selection like that in feb for 'boot to init.sh', 'run upstart with debug' etc
[14:43] <wookey> I have 'kcmt' kernel command line modification tool' here, which maybe helpful
[14:44] <wookey> I assume we are using a dtb in the above .axf files?
[14:45] <suihkulokki> oops right, the dtb file inside the axf is different too
[14:46] <wookey> OK. I'll have a poke abut with this - looks like it should be kickable
[15:14] <infinity> wookey: That sounds less like it's hung and more like you don't have a getty on your serial console.
[15:14] <infinity> wookey: Though, that last message from init is a bit odd-sounding, I suppose.
[15:19] <wookey> Ah OK, that's possible
[15:21] <wookey> I guess that's /etc/init/console.conf in upstart-world. which just runs exec /sbin/getty -8 38400 console
[15:23] <wookey> which is just the same as my raring img
[15:33] <infinity> wookey: If this is on a machine/vm with an actual serial console, you'll want an /etc/init/ttyAMA0.conf or ttyS0.conf or whatever.
[15:34] <wookey> OK. there isn;t one in the working image
[15:34] <infinity> wookey: Like http://paste.ubuntu.com/6283597/ for instance.
[15:34] <infinity> Though, if /dev/console is actually the right thing, then nevermind.
[15:34] <infinity> Is this for a Foundation Model?
[15:37] <wookey> yes
[15:37] <wookey> I'm trying to fish the options out of the working axf - thenI might have some idea what's what
[15:44] <infinity> I believe apw got ubuntu-core booting on a Foundation Model the other day in a fit of experimentation and boredom.  Not sure if he had to do anything fancy to get there.
[15:44] <infinity> apw: Or am I on crack?
[15:48] <apw> infinity, indeed, i'm doccing it at the moment
[18:15] <wookey> OK. I've grokked how to get kernel command lines in and out of .axfs
[18:16] <wookey> and Iwas just about to complain that it still hansg with upstart, but after impressive wait I have a prompt :-)
[18:17] <wookey> I appear to be in business
[18:17] <wookey> OK. so I guess the conclusion is that saucy doesn;t boot with 3.8.0 on vexpress, but does with 3.11
[18:18] <wookey> The secret arm internal tool modify_commandline.sh ( load of objdump and dtc runes) ought to get a wider airing - it's exceedingly handy for this sort of thing
[21:17] <AmEv> Any Tegra 2 guys here right now?
[21:55] <wookey> infinity, doko: so I booted http://cdimage.ubuntu.com/ubuntu-core/releases/13.10/release/ubuntu-core-13.10-core-arm64.tar.gz but any attempt to use apt just complains that libstcd++ is not present. Is that expected?
[21:56] <wookey> it does seem not to be present
[21:57] <wookey> presenably if I jam http://ports.ubuntu.com/pool/main/g/gcc-4.8/libstdc++6_4.8.2-1ubuntu2_arm64.deb in things should work better
[22:03] <AmEv> Is there a porting guide to get Ubuntu onto Android tablets natively? (deep wishing.... haha....)
[22:04] <AmEv> I've tried following several Tegra 2 guides, but never had luck booting my tablet into Ubuntu.
[22:05] <doko> wookey, I never did see that
[22:07] <doko> I'll copy the tarball and image which I am using, but it will take until tomorrow
[22:08] <wookey> OK. cheers.
[22:09] <wookey> hmm in fact dpkg says libstdc++6:arm64 is installed.
[22:10] <wookey> but apt-get update says: pt-get: error while loading shared libraries: libstdc++.so.6: cannot open shared object file
[22:11] <wookey> aha /usr/lib/aarch64-linux-gnu/libstc++.so.6 is a link to /usr/aarch64-linux-gnu/lib/libstdc++.so.6 which is missing
[22:13] <wookey> looks like toolchain build issue
[22:15] <doko> not here
[22:18] <infinity> wookey: I tested it and didn't have that issue.  Did you do something curious to it?
[22:21] <wookey> nope, just unpacked
[22:22] <wookey> what version of libstdc++6 have you got?
[22:23] <wookey> 4.7.3-1ubuntu1 here. The 4.7.3-1ubuntu2 on ports seems to have a real /usr/lib/aarch64-linux-gnu/libstdc++.so.6.0.18 in it
[22:23] <infinity> (base)adconrad@cthulhu:~/core$ ls usr/lib/aarch64-linux-gnu/libstdc++.so.6* -l
[22:23] <infinity> lrwxrwxrwx 1 root root     19 Oct 11 20:16 usr/lib/aarch64-linux-gnu/libstdc++.so.6 -> libstdc++.so.6.0.18
[22:23] <infinity> -rw-r--r-- 1 root root 975016 Oct 11 20:27 usr/lib/aarch64-linux-gnu/libstdc++.so.6.0.18
[22:23] <infinity> wookey: ^-- I see nothing wrong with the above, so I'm questioning how you unpacked it.
[22:24] <infinity> (That was a fresh unpack of a wget of the above URL you gave)
[22:24] <wookey> hmm. I have got those files
[22:25] <infinity> Your symlink sounds suspiciously like it would have come from a cross toolchain.  Which is impossible in the core tarball itself.
[22:25] <doko> err, 4.7.3 ???
[22:25] <infinity> wookey: http://paste.ubuntu.com/6285690/
[22:26] <infinity> wookey: Whatever you've done, it didn't come from my tarball. :P
[22:27] <wookey> I really did just do: tar -C $MOUNT -xzf $TARBALL
[22:27] <wookey> (inside losetup mount to make a .img disk image)
[22:28] <infinity> And there was nothing previously existing in said directory?
[22:31] <wookey> it gets formatted then mounted - looks farily robust. I'll poke about a bit and see what's what
[22:33] <doko> wookey, see http://people.canonical.com/~doko/tmp/arm64/ this one has a saucy host, and a trusty chroot, my sync will need about on hour
[22:33] <infinity> wookey: And so you don't think I'm crazy, on real hardware:
[22:33] <infinity> http://paste.ubuntu.com/6285718/
[22:34] <wookey> what do you get for  dpkg -l | grep libstdc ?
[22:34] <infinity> wookey: That's at the end of that paste.
[22:35] <wookey> right. OK. I'm clearly not running what I think I'm running.
[22:36] <wookey> thanks for the reality check
[22:36] <infinity> wookey: Well, I'd have noticed if this was broken, given that "chroot in and run apt-get a bit" is basically *the* test case for ubuntu-core validation. :)
[22:37] <wookey> indeed. no apt is serious
[22:49] <wookey> OK. obvious and embarassing s/raring/saucy/ error spotted. trying again...
[23:25] <doko> wookey, sync finished, http://people.canonical.com/~doko/tmp/arm64/