=== jrgifford is now known as Guest87195 === zz_chihchun is now known as chihchun === chihchun is now known as zz_chihchun === zz_chihchun is now known as chihchun === doko_ is now known as doko [07:19] good morning [08:40] hello === yofel_ is now known as yofel [09:33] can anyone reply to https://twitter.com/yuvilio/statuses/273566993171492864 please? [09:40] dholbach: It should be roughly the same, hardware-wise, as a Chromebook, if that's helpful. [09:41] right [09:41] dholbach: There's kernel work and hobbyist fun going on there, but not by us. [09:41] * ogra_ isnt aware of any official statements about the nx10 though [09:42] i guess a chromebook kernel might show up in the archive before release [09:42] so someone who would try to get it to work should be able to [09:42] I'm still tossing around the idea of getting an American to send me a Chromebook. [09:42] Which would give me incentive to maintain a kernel in the archive. :P [09:42] (cant be to much difference between the two) [09:43] Part of me is still holding out for Lenovo to make an ARM laptop, so I don't have to compromise on my love of Lenovo keyboards. [09:43] But I could be waiting a long time for that... [09:44] n10 kernel is...rather different for some reason., [09:44] they don't use DTB's [09:44] but other than that, it should be trivial. [10:04] thanks everyone [10:30] dholbach, oh, and a sidenote, no redistributable GLES drivers for the nexus 10, unless that exists its very unlikely for us to switch to it simply because we cant run unity [10:30] http://marcin.juszkiewicz.com.pl/2012/11/28/lets-compare-some-cpu/ - you may find it interesting [10:31] ogra-cb, if the guy replies I'll let him know :) [10:31] k [10:31] that could become our general answer to that faq :) [10:31] (i hear that question once a week, but ignore it most of the time ... ) [10:33] ogra-cb: n10 has same problem as chromebook [10:34] yes [10:34] but has no chromeos to hack around the issue [10:34] ogra-cb: if you can spare 363MB of data and 1.4GB of disk space then I have solution [10:34] on the chromebook you can at least pull the libs from chromeos [10:34] s/data/network data/ [10:34] cant do that on the nexus10 [10:35] you have a 363MB download that solves the licensing ? [10:36] chromebook recovery image can be fetched [10:36] but user would have to do it ;( [10:37] the real pita of exynos5 is lack of any license infromation in chromium os (or I did not yet found it) [10:38] https://dl.google.com/dl/edgedl/chromeos/recovery/chromeos_2913.84.10_daisy_recovery_stable-channel_mp-v2.bin.zip [10:38] fetch, unpack, use kpartx to split partitions, mount ROOT-A as ro, copy files [10:39] 363MB zip, 1GB img [10:39] http://commondatastorage.googleapis.com/chromeos-localmirror is also useful if you know what you are searching for. [10:39] but no libmali there === lardman|work is now known as lardman [10:54] hrw: libmali is the userspace closed lib? [10:54] yes [10:54] lardman: long time - how things? [10:55] hrw: indeed, not bad, 14month old daughter has sort have put a hold on hobby coding up till now [10:55] but now we're getting at least some sleep, I'm looking for some hw :) [10:55] lardman: ;) [10:55] 14m should be able to sleep whole night I think [10:56] hrw: If you'd like to tell her that I'd be happy ;) [10:56] lardman: ;)) [10:57] So back on topic, are there blobs for the Mali 400? [10:58] no idea of mali400 [10:58] you have exynos4 or allwinner a10? [10:58] nothing yet, umming and arring about the Nexus 10 [10:59] nexus10 has mali t604 [10:59] alternatively the Nexus 7, but I'd quite like to do some on-device opencl (if an sdk is released eventually) [10:59] yeah, was just wondering whether they are just running slow releasing the relevant files [10:59] I would not assume that Samsung will release anything [10:59] the opencl sdk should come from ARM I'd have thought [11:11] howdy lardman [11:11] hey lilstevie [11:11] am heading over to your neck of the woods early next year [11:12] well not necessarily that close, but closer than here anyway [11:12] heh [11:12] where abouts? [11:12] Adelaide [11:13] to visit family [11:13] ah fair enough [11:13] not close, but closer :p [11:13] yep [11:20] hmm, interesting that the only installer for the the latest Mali OpenGL ES SDK for Linux on ARM is a Windows MSI file... :) [11:22] nothing interesting in there though, shame [11:30] xnox, did you btw research why the current /usr/lib/ubiquity/wallpaper fails or did you only look at compiz integration yet ? [11:30] ogra_: working at compiz bits right now. Did not troubleshoot the wallpaper fail yet. [11:49] xnox, ah, finally found the bug again (and gave it a better title) bug 1081260 [11:49] Launchpad bug 1081260 in ubiquity (Ubuntu) "/usr/lib/ubiquty/wallpaper crashes after boot of livecd" [Undecided,Confirmed] https://launchpad.net/bugs/1081260 [11:50] ogra_: well it's interesting, but the "crash" reproducer is wrong as it needs to be called with the path to background (and error message says so) [11:50] yeah [11:50] ogra_: but yeah, I'm in gtk right now to fix a compiz bug ;-) [11:50] http://paste.ubuntu.com/1394259/ [11:50] thats his actual error in the ubiquity-dm log [11:51] seems also that wallpaper.c still uses gtk2 [12:04] bug 1048976 [12:04] Launchpad bug 1048976 in ubuntu-wallpapers (Ubuntu) "[UIFe] Update default ubuntu wallpaper to quantal version" [Undecided,Fix released] https://launchpad.net/bugs/1048976 [12:26] oha ! [12:26] plymouth works if i enable debugging [12:27] ogra-cb: which kernel argument? [12:27] plymouth:debug [12:27] cool [12:27] btw - what plymouth does other than bootsplash? cause I do not see any difference when boot chromebook without plymouth [12:28] well, i also have break=bottom enabled might be that there is a race i avoid through that [12:28] bootslpash is an in kernel hack [12:28] ? [12:28] thats the only "bootsplash" i know [12:29] ok, let me rephrase... what plymouth does at all? [12:29] hmm, dropping plymouth:debug still gives me a splash [12:30] hrw, libplymouth does all kernel-userspace interaction (like a poor mans dbus for UI stuff, i.e. passwords, fsck input etc) [12:30] thats why i.e. mountall requires it [12:30] the rest is just bling [12:31] hmm, so am i brave and try it without break=bottom now ... [12:31] at the risk of having to reinstall my nexus [12:31] thx [12:32] * ogra_ doesnt get why it breaks at install time so badly === chihchun is now known as zz_chihchun [12:34] aha ! [12:34] so it crashes id i dont have either plymouth:debug or break=bottom set [12:39] http://code.google.com/p/chromium-os/issues/detail?id=36717 [12:40] ++ [12:41] will fill similar ones for bunch of others [12:41] openmax, mfc firmware [12:41] openmax, yeah [12:41] ogra_: http://commondatastorage.googleapis.com/chromeos-localmirror/distfiles/openmax-0.0.1-r5.tbz2 [12:42] ogra_: binaries only [12:42] do they work ? [12:42] ogra_: no experience with openmax here [12:43] i only have experience with gst-omx [12:43] that doesnt look like it would be anyhow related to gst [12:43] yep [12:46] ogra_: both ubiquity panel & wallpaper are gtk3. [12:47] ok [12:48] i thought i saw a gtk2 header but seems i looked at precise code here [12:48] (sorry for the false alarm due to that) [12:51] ogra_: interesting bug 296538 [12:51] Launchpad bug 296538 in unity-2d "warty-final-ubuntu.png is actually a JPEG file" [Undecided,In progress] https://launchpad.net/bugs/296538 [12:51] ogra_: /me ponders if the wallpaper works on e.g. xubuntu. [12:51] a bit old though [12:52] that wouldnt explsin it on the nexus7 though [12:52] *explain [12:52] we use the default ubuntu-wallpapers wallpaper [12:52] ogra_: red-herring or something got strict. [12:52] * xnox sees no wallpaper in the amd64 VM [12:53] ah [12:53] so not arch or flavour specific [12:54] kvm that is. [12:55] anyway, fixed the compiz bug will do a merge proposal now =)))) [12:56] ogra_: compiz is the last fallback, so on the images you'll need to somehow unseed metacity. [12:57] what i still wonder is why g-s-d doesnt just default to draw the wallpaper [12:57] it should just use the system defaults the desktop uses later too [13:00] * ogra_ files bug 1084063 [13:00] Launchpad bug 1084063 in plymouth (Ubuntu) "plymouth in raring seems to have a race condition on the nexus7" [High,New] https://launchpad.net/bugs/1084063 [13:00] i have not the slightest idea how to debug this :( [13:00] the kernel goes into a reboot loop so i wont be able to capture any info [13:24] xnox, oh ! if i can unseed metacity, i can re-seed the slideshow i guess [13:24] without it eating all my space [13:25] ogra_: well merge proposal is up. As soon as it's reviewed, I'll merge and upload new ubiquity ;-) [13:27] xnox, approved ;) [13:27] xnox, i'm wondering though if we couldnt merge all these /proc/cmdline parsing into one function and set all vars we need at the top [13:28] ogra_: cheat. what about three-pairs of eyeballs? =) [13:28] well, if you feel like [13:28] i assume you have tested that before requesting [13:28] ogra_: also on nexus do you need to move the ubiquity because of onboard? [13:29] move ? [13:29] ogra_: yes, literarly move the ubiquity window around the desktop [13:29] the current nexus image is fine apart from missing slideshow and wallpaper [13:29] * xnox didn't enable that compiz plugin, nor resize. [13:29] ack. [13:29] ok, merge and upload I guess =) [13:29] onboard cares for that itself [13:29] cool =) [13:30] if it detects it covers the focused input area *it* moves around [13:30] no need to move the windows ;) [13:30] its a bit jumpy in the user setup screen [13:30] but nothing we can avoid atm [13:31] unless we switch to ion3 or awesome and use a fixed two pane layout ;) [13:33] * xnox glares at ogra for daring to mention tiling window managers [13:33] hey its a tablet, they are perfect for that [13:34] its the predecessor to BathroomWindows^WWindows Metro === zz_chihchun is now known as chihchun [14:01] * ogra_ adds various sleeps to the initrd init to see if that helps plymouth [14:01] hmm, no, doesnt look like [14:15] YAY ! [14:19] * ogra_ updates bug 1084063 [14:19] Launchpad bug 1084063 in plymouth (Ubuntu) "plymouth in raring seems to have a race condition on the nexus7" [High,New] https://launchpad.net/bugs/1084063 === albert is now known as Guest72822 === Guest72822 is now known as Phryq === albert_ is now known as Phryq [16:27] bug 1083723 [16:27] Launchpad bug 1083723 in upstart (Ubuntu Raring) "'telinit u' has a cage fight with busybox init" [Medium,Triaged] https://launchpad.net/bugs/1083723 === chihchun is now known as zz_chihchun === rickspencer3_ is now known as rickspencer3 [18:01] ogra_: are the new dailes down in size to not have to use -S? [18:08] vanhoof, yes, they should just work now (limited to 6G indeed) [18:08] ogra-cb: sweet [18:09] i havent tested todazs, but yesterdays was fine [18:09] *todays [20:14] <[mbm]> did anyone ever get anywhere with nexus7 multiboot? [20:14] yeah) [20:15] <[mbm]> link? [20:15] http://forum.xda-developers.com/showthread.php?t=2011403 === jrgifford is now known as Guest13203 === Todd__ is now known as Todd [23:52] ogra_: can we purge https://wiki.ubuntu.com/ARM/RootfsFromScratch/QemuDebootstrap, https://wiki.ubuntu.com/ARM/RootfsFromScratch, https://wiki.ubuntu.com/ARM/RootStock from the face of earth [23:52] ogra_: and redirect people to: $ ubuntu-core images or $ mk-sbuild --arch armhf raring [23:53] ogra_: do we have cloud-init armhf images? those work great with quick download & launch a qemu instance.