[00:08] how the daily changelog will look: http://paste.ubuntu.com/5615219/, should be around tomorrow [00:08] will it be on cdi site? [00:09] yup [00:09] comparing with the previous build [00:09] are we able to get a raring build? [00:09] not yet, in progress [00:10] by progress I mean some packages are being pushed [00:10] once we have enough (the most critical ones), we can do the switch [00:10] and hopefully remove the need of a ppa as well [00:10] gotcha === XenGi is now known as XenGi_ [00:11] will daily builds soon include daily-core-apps? [00:11] rsalveti, where it's possible to watch ubuntutpuch changelog ? [00:12] RobbyF: yup, soon (I also want them) :-) [00:12] http://cdimage.ubuntu.com/ubuntu-touch-preview/ should be in here I believe. [00:12] krabador: it'll be available together with the images === XenGi_ is now known as XenGi === XenGi is now known as XenGi_ [00:31] Trying to deploy an app to Ubuntu Touch on my Nexus 7. It's prompting for a password but not accepting the default "phablet" Anyone else run into this? [00:33] how are you installing? [00:34] RobbyF: I am using QTcreator. It is seeing the device, and I can open ssh to it using the default password. but "Run on Device" or ctrl+f12 promps for a password. The "phablet" password works for the ssh connection but not the deploy to device connection [00:35] hmm. that i'm not sure of. sorry. === XenGi_ is now known as XenGi === XenGi is now known as XenGi_ === XenGi_ is now known as XenGi [00:47] it's may be YOUR computers root password [00:48] I just reported Launchpad bug 1155382 in qt-folks "[crash] QContactManager is accessed after deletion" [Undecided,New] https://bugs.launchpad.net/qt-folks/+bug/1155382 [00:48] Launchpad bug 1155382 in qt-folks "[crash] QContactManager is accessed after deletion" [Undecided,New] [00:53] rsalveti, hey, I'm trying to build qtvideo-node, but I can't seem to install two of the build deps (libhybris-dev and libqtubuntu-media-signals-dev) is there a special ppa for those that I'm missing? [00:57] robru: the main ppa is https://launchpad.net/~phablet-team/+archive/ppa [00:57] which you should find the needed packages [00:58] camera app? [00:58] rsalveti, thanks... I see libhybris there but not libhybris-dev. is there such a thing as libhybris-dev? perhaps qtvideo-node has broken build-deps... [00:59] is that real? [00:59] robru: yup, at this view you can only find the source packages [00:59] at https://launchpad.net/~phablet-team/+archive/ppa/+packages you can find the packages that hybris produces [00:59] just open the tab [01:00] rsalveti, hmmm, that PPA warns that it will break my desktop! how am I supposed to build qtvideo-node... :-/ === XenGi is now known as XenGi_ [01:00] robru: right, you could build in a chroot, but still, problem is that the result will only be valid for the device (if you build for armhf) [01:01] maybe qtvideo-node shouldn't depend on hybris at the desktop use case (i386, for example) [01:02] rsalveti, yeah, I'm not expecting to be able to run phablet stuff on my desktop; didrocks simply asked me to investigate some of the packaging of some of the phablet stuff, so I'm just trying to do a basic 'bzr bd' to make sure that stuff is buildable and correct... but I can't build qtvideo-node without libhybris-dev... so should I install libhybris-dev on my desktop or not? (i have no idea what it is) [01:03] robru: libhybris is the package responsible for doing the symbol loading for the android libraries, it's useless for your host [01:03] you can install libhybris at your host, the problem with that ppa is if you enable and install some other packages [01:04] bbl [01:11] wow [01:13] ?? [01:14] hi [01:14] whats up [01:18] Just finished watching the Sam galaxy S4 event === paradoxgo- is now known as paradoxgo [01:25] RobbyF it was a sad horrifying cacophany IMO [01:26] I really liked it [01:27] better than the traditional events. [01:27] yeah [01:33] say I if your here and alive [01:33] I [01:34] I [01:35] J [01:35] close enough === juicyjones is now known as juicyjones|away === juicyjones|away is now known as juicyjones === rukin5197|afk is now known as rukin5197 [02:42] hi [02:43] what is the ppa that has the core apps? I forgot, and I just reflashed my Nexus4... [02:45] ppa:ubuntu-touch-coreapps-drivers/daily [02:55] RobbyF: thanks. But I just tried to instal them on the ubuntu touch device, did not work (did an apt-get update prior to install). Shouldn't it? [02:56] no [02:56] well ya [02:56] what did you try to install [02:56] sudo apt-get install ubuntu-calculator-app [02:57] apt-get install ubuntu-calculator-app ubuntu-calendar-app etc... [02:57] it will install them [02:57] but you need to run them from ssh/console still. [02:58] I'm not sure the correct commands to do that though. [02:58] never mind. I installed software-properties-common but never added the ppa :P [02:59] I don't think you need to install that on the latest build. [02:59] hmm [03:00] how do we run the app? [03:02] it seems we need to desktop_file_hint parameter [03:02] anyway [03:02] I added the repository and now it worked, I am going to install the apps [03:02] there is a way to launch [03:02] just not locally [03:04] any d2att/i747 user [03:04] * dank101 needs a tester [03:11] anyone? [03:15] not me [03:15] ok [03:15] my tester got back === XenGi_ is now known as XenGi === XenGi is now known as XenGi_ [03:54] look under devices [03:54] https://wiki.ubuntu.com/Touch/Devices# [03:54] search for AT&T === crazydude101 is now known as RamchandraApte === resba|Off is now known as resba [04:36] why do my qmlscene windows not appear in google hangout screenshare list? [04:46] has anyone else tried that? === freeflyi1g is now known as freeflying === Lloir|work is now known as Lloir|zZzZzZz === mmrazik is now known as mmrazik|afk [06:15] hi === juicyjones is now known as juicyjones|away === mmrazik|afk is now known as mmrazik [06:49] ok so i just got this error [06:49] make: *** No rule to make target `bacon'. Stop. [06:49] how do i fix this? [06:51] rsalveti: do you know who could refresh qtwebkit's add_experimentalDevicePixelRatio.patch patch on top of qtwebkit 5.0.1? [07:29] good morning === IanWizard-Cloud is now known as IanWizard-Person [07:40] dholbach, good morning :) [07:40] hey tvoss [07:40] hey dholbach [07:41] dholbach, wanna give me a call? [07:42] tvoss, I need to finish a few small things - how about in 30m? [07:42] dholbach, wfm [07:42] perfect! === IanWizard-Person is now known as IanWizard-Cloud [07:58] hey. i'm trying to make a port of ubuntu touch for galaxy ace plus, but i don't have any cm.dependencies file, and can't understand what to do with the Android blobs === chriadam is now known as chriadam|away [08:48] dholbach, wanna jump over to #ubuntu-mir? === davidcalle_ is now known as davidcalle === yofel_ is now known as yofel [10:13] kaleo: so qt5-beta-proper has mostly Qt 5.0.1 identical packages for precise and quantal as raring has [10:14] the good question is when are touch image people starting to do builds against it, since I assume I can't simply copy those packages over to qt5-proper before some build testing has been done etc [10:14] Mirv, though I'm wondering, is it a lot of work to backport the fix for SVG support? [10:15] Kaleo_: not really, I just need to test build it first in some 3rd/4th repo, and then copy it over there. can be done. [10:15] backporting the packaging to the 5.0.0 sources [10:16] kaleo: putting that to to-do-soon [10:16] (as in today) [10:16] Mirv, great! [10:16] Mirv, designers will be happy to hear that [10:23] Mirv: my sdk still doesn't work i think: http://imagebin.org/250357 [10:23] or is that how it's supposed to look? [10:28] Just got the "Device detected as tilapia. Unsupported device, autodetect fails device" error installing to Nexus 7 3G [10:29] resinstalling simmulating device as "grouper" instead of "tilapia" [10:29] ... hope this is compatible - anyone have experience [10:34] Skilly you have openjdk6 or 7 installed to your workstation? [10:34] yea - both [10:35] I got very odd error messages without the jdk's installed and those are not in the phaplet dependency files. Ok, then no issues, cannot help u :( [10:35] ali1234: looks similar to mine [10:36] thanks dun1982. Seems tilapia is not yet supported [10:36] there might be changes brewing though, the plugin is very new [10:36] quite widely reported via google - impersonating grouper is the reported workaround [10:38] Mirv: shouldn't there be special mobile app templates? and target <> desktop? [10:46] are the videos from yesterday up yet? [10:51] the recordings of the SDK days can be found on https://www.youtube.com/user/UbuntuOnAir [10:56] jppiiroi1en: can you answer ali1234 about those? [11:02] ali1234, Mirv: http://www.youtube.com/user/UbuntuOnAir [11:11] jppiiroi1en: I mean the templates vs mobile thing :) [11:12] trying to see in the video :) [11:13] oh he doesn't show it :( [11:14] urgh [11:17] ali1234: http://www.youtube.com/watch?v=JCOspJaZmpo [11:18] maybe that answers [11:24] Great news! Because of a goddamn wind gust I dropped my Nexus 7 and now have a lot of small Nexus 7s ... [11:24] lol [11:24] Yea, 'cept it's not that funny from my angle. [11:25] More like 200€ blown into the wind ... literally. [11:25] I know, just the way you said it that was... [11:25] No insurance? [11:25] Sucks [11:26] It was a used one, imported from Canada and I'm in Austria. [11:26] I hate the nature right now. [11:31] mainerror is that an excuse to upgrade when the next one comes out in may [11:31] Not really ... bought mine only three months ago. [11:32] heh ok frustrating [11:33] i''m reluctant to buy top end mobile hardware eg laptops for this sort of reason. i have ended up liquid-damaging 2 laptops [11:33] i'm clumsy [11:33] First time ever for me. [11:34] If it wasn't for this 70km/h wind we have right now it would still be alive ... [11:35] Anyway ... what happened that happened. It's history. [11:36] Does canonical plan to make the whole ubuntu phone stack, including end-user applications, free software? Or will some components be closed? [11:36] (I know they will not have authority over the kernel) === jpetersen_ is now known as jpetersen [12:04] ah ok, looks like my qt creator is working correctly === MacSlow is now known as MacSlow|lunch [12:25] hmm, i remember someone saying there's a port of EGL hello_triangle? [12:25] any idea where i can find that? === tvoss is now known as tvoss|lunch === mmrazik is now known as mmrazik|lunch === mmrazik|lunch is now known as mmrazik === _salem is now known as salem_ [12:54] kaleo: I wonder if you can briefly somehow test the new svg package before I copy it to qt5-proper? [12:57] rsalveti, https://code.launchpad.net/~ogra/session-manager/posix-rewrite/+merge/153542 ... happy reviewing :) === juicyjones|away is now known as juicyjones === XenGi_ is now known as XenGi [13:32] ok, discussed the Qt SVG thing and copied it over to qt5-proper PPA [13:32] ie. it missed the support libraries, and I backported the fix to the 5.0.0 package === MacSlow|lunch is now known as MacSlow === tvoss|lunch is now known as tvoss [14:12] Hello everyone :) I'm new in the world of Launchpad and I want to understand how it works. I did a little bugfix (a stupid typo), then I committed the change, used dch -i to modify the changelog, pushed the branch and propose for merge, but they said me [14:12] "Actually sorry, probably want to keep the changelog as UNRELEASED rather than targeting precise" (i'm on precise) [14:13] How have i to modify changelog? [14:13] Thanks for everyone that help me, and sorry for bad english! [14:13] https://code.launchpad.net/~rpadovani/phablet-tools/fix-for-1139999/+merge/153419 [14:20] tvoss, ping [14:25] paulk-desktop, pong [14:26] tvoss, I was wondering whether Ubuntu Touch would run fast enough without the non-free vendor blobs for graphics (gralloc, hwcomposer, EGL libs) [14:27] paulk-desktop, I think the better question is: would it run at all? we do not have fallback software implementations of those components right now [14:27] there are software implementations for all of these in Android [14:27] so you most likely have them too [14:28] paulk-desktop, if they are readily usable, it is worth trying [14:28] paulk-desktop, but I haven't checked myself on them [14:28] ok [14:28] paulk-desktop, can you give me a pointer to those implementations? [14:28] tvoss, hardware/libhardware/modules/ [14:28] and sw EGL is in frameworks/base/ [14:29] frameworks/base/opengl/libagl [14:29] and frameworks/base/opengl/libs [14:29] there must be at least one video card where that stuff isn't proprietary [14:30] ali1234, wanna bet? [14:30] software implementation on arm isn't going to run well though [14:30] ali1234, it does run well on Android === Lloir|zZzZzZz is now known as Lloir [14:30] it might do if you just tell Qt to directly use it's own software rendering [14:30] ok [14:30] no-go on qtquick2 at least [14:30] but software implementation at the very low levels is super slow [14:31] just try running unity on llvm-pipe... it blows [14:31] ok [14:31] but unity-2d is fine and does pretty much the same thing [14:31] just at higher level [14:31] paulk-desktop, what problem are you trying to solve? [14:32] tvoss, the problem of proprietary software :) [14:32] I was just wondering whether you tried it already or not [14:32] so I guess it'll up to me for testing [14:32] paulk-desktop, nope, we haven't looked into it [14:32] (I guess I'll wait until Ubuntu Touch becomes more clean and usable though) [14:33] tvoss, thanks for your answers then [14:34] paulk-desktop, I'm looking for the opengl implementation right now, but cannot find it in that directory [14:34] paulk-desktop, yw [14:34] tvoss, I'm pretty sure it's already in your system [14:34] that's the 0 0 android config in egl.cfg [14:34] (which is most likely missing if you are based on CyanogenMod) [14:35] paulk-desktop, looking at aosp right now, but I might be missing something [14:35] paulk-desktop, got it, thanks [14:35] but basically, if you wanna try, you can push that only line in egl.cfg (and remove the other one) and remove the vendor gralloc/hwcomposer [14:35] that should do it [14:37] okey, back [14:39] matge: hey, I have a branch with the stuff we talked about but I havent pushed it anywhere, I'll set something up on github later [14:39] like keyboard docking and reflow of views with it [14:43] ricmm: that's cool! [14:46] kaleo, ping [14:53] tvoss, pong [14:53] rsalveti: I'm just realizing: could we make phablet-tools native instead of worrying about packaging changes vs. upstream changes and rolling tarballs and such? [14:53] Kaleo_, you kaleo or his evil twin? :) [14:53] tvoss, I'm both :) [14:54] * tvoss hides away [14:54] lool: sure, but why the concern? [14:55] lool, and while you're at it can you upload it to the archive too ? :) [14:55] guess we'd need some more work before it hit the archive [14:55] lool, ogra_: better, get it to be added to the autolanding stack! [14:55] move to python 3, for example [14:56] seb128, as long as the end product lands in the archive i dont care how it got there :) [14:56] rsalveti: when I did some uploads just before the mwc demo images went out, I did changes in the bzr branches which were considered upstream, but which got included in the .diff because there was an upstream tarball; eventually some time later Sergio rolled a new tarball; I found this to be unnecessarily complicated; similarly there's a merge proposal from another Ricardo to fix a typo, which is in the "upstream" code but uses a non-upstream version [14:57] rsalveti: just having to worry about upstream vs. non-upstream changes feels weird here since the package essentially lives in Ubuntu right now [14:57] Mirv: I can rebase it, though I would really like to not need it [14:57] it just crossed my mind that it ought to just be native [14:57] seb128: yeah, that's a good point too [14:57] seb128: how are native packages handled in autolanding? [14:58] yeah, don't remember if it was a recommendation to not use native [14:58] rsalveti: I guess it's a case of pinging Sergio when he's back? [14:58] lool, "well"? [14:59] the recommended source format was 1, not native [14:59] seb128: like the place you're throwing coins and make a wish? [14:59] lol [14:59] guess seb128 might know more as well [14:59] he seems to know a lot about fetching water in the well [15:00] lool, does native/non-native makes any difference? [15:00] for daily landing you basically get a new tarball from the vcs for each landing [15:00] seb128: upstream tarball vs. lack of one [15:00] if we always bump the upstream version, it makes no difference [15:00] ok [15:00] when creating new release [15:00] rsalveti: let's discuss this with Didier in 1h [15:00] the daily landing does a "release" every day [15:01] yeah [15:04] ogra_: did you see the changelog email? [15:05] rsalveti, ah it drowned in the other mails, seeing it now .... and fioxing the sync script :) [15:06] ogra_: cool, thanks :-) [15:06] rsalveti, did you see the merge request ? :) [15:06] (its huge do it if you have some spare time) [15:06] ogra_: in my list [15:06] awesome [15:06] i'll take a hit on the other bash script in that branch too over the weekend [15:07] though thats using even more arrays ... might take a bit more [15:07] ogra_: right, cool [15:07] slangasek, I was already here in fact. Got lost in my channels [15:07] but it will have a dircet impact on startup speed [15:07] yeah, it's painful [15:07] Mirv: I suppose you want the package updated at the 5.0.1 based ppa? [15:08] wonder if this should also be pushed to the archive [15:10] slangasek, so I have a WI to get qtubuntu in shape for autolanding, but it's only armhf because it requires android platform bits. Are you familiar with the technical details there? I'd be happy to help adjust it to work on i386 and amd64, but I'm not sure yet what bits it's relying on and thought I'd ask before diving into the code [15:11] Has there been discussion on how to do the online account plugins on the phone? They are now gtk based, which does not work currently === salem_ is now known as _salem [15:20] Does anyone know if USB host mode is working on the Nexus 4 with the current version of Ubuntu Touch? [15:22] MrAdam: I read that usb host mode is not supported by the hardware [15:23] Oh well, that settles it then ^_^ [15:23] Thanks [15:23] MrAdam: but there are hardware hacks supplying the necessary voltage externally [15:23] Mmh.. My old Galaxy Nexus had it. I just assumed the Nexus 4 did as well [15:24] galaxy nexus doesn't have usb host mode [15:24] afaik [15:24] it only uses mtp [15:25] It does indeed have host mode [15:25] I used it in a few apps I've developed [15:25] It was also possible to connect a USB drive to it and mount it === greyback is now known as bzoltan-smells === bzoltan-smells is now known as greyback [15:34] PLEASE NOTE: We're in the first session of https://wiki.ubuntu.com/UbuntuSDKDays [15:34] you can watch it live on http://ubuntuonair.com/ [15:34] and ask your questions in #ubuntu-app-devel === Ursinha_ is now known as Ursinha === rsalveti_ is now known as rsalveti [15:37] rsalveti: yes, and I already did with a very-certainly-not-perfect version of the patch, but I'd like to have a patch that is properly adjusted to apply on top of https://launchpad.net/~canonical-qt5-edgers/+archive/qt5-beta-proper/+files/qtwebkit-opensource-src_5.0.1.orig.tar.xz [15:37] rsalveti: what gets done in that packaging branch gets eventually pushed to archives as well [15:38] Mirv: alright, need to disable jit for arm as well I'd say, I'll review the patch/package later today [15:38] to make sure it's in sync with our own === dandrader is now known as dandrader|afk [15:40] rsalveti: I did that, although I haven't tested if it works [15:40] Mirv: ok, cool [15:40] rsalveti: lp:~kubuntu-packagers/kubuntu-packaging/qtwebkit-opensource-src [15:40] so the device pixel ratio patch is mangled, should be replaced, and the disabling optimizations I modified to be arm only === _salem is now known as salem_ === dandrader|afk is now known as dandrader === guest1__ is now known as guest1_ [16:29] kenvandine, libdee-qt5-dev does not seem to ship a .pc file anymore (looking at version 3.0-0ubuntu2~phablet1 in phablet-team/ppa/ubuntu) [16:29] Saviq, ^^ [16:30] Kaleo_, it does in my package in raring [16:30] Kaleo_, hum, checking [16:30] kenvandine, I forgot to say, on armhf [16:30] Kaleo_, Saviq just noticed yesterday that the lib is wrong [16:30] kenvandine, sorry, forget that comment, it's there; I'm blind [16:30] kenvandine, something else is wrong somehow [16:31] the package in the phablet PPA is broken, we think [16:31] kenvandine, ah yes on my amd64 laptop: /usr/lib/x86_64-linux-gnu/pkgconfig/libdee-qt5-3.pc [16:31] kenvandine, on the phone: /usr/lib/arm-linux-gnueabihf/pkgconfig/libdee-qt5.pc [16:31] no -3 [16:32] tsdgeos, actually that might need another fix before the shell release ^ [16:32] kenvandine, I have version 3.0-0ubuntu1phablet5 on my laptop (probably older) [16:32] Kaleo_, btw, https://launchpad.net/~phablet-team/+archive/ppa/+sourcepub/3028305/+listing-archive-extra is building with the fixed libname [16:33] "sudo mv /usr/lib/arm-linux-gnueabihf/pkgconfig/libdee-qt5.pc /usr/lib/arm-linux-gnueabihf/pkgconfig/libdee-qt5-3.pc" fixes it for me :) === salem_ is now known as _salem [16:36] kenvandine, btw, you tricksed us with your dee-qt package in super-friends ;P === juicyjones is now known as juicyjones|away [16:36] woops :) [16:36] kenvandine, was > version than our phablet one ;) [16:37] sorry :) [16:37] kenvandine, all fixed now [16:37] although it is less broken :) [16:37] indeed [16:37] i guess that forces some rebuilds ;) [16:37] i've been trying to get everything in there so app developers can play after my SDK dev days session today [16:37] Saviq: why do we have our own phablet branch for dee btw? [16:38] tsdgeos, it has the deevarianttext merged [16:38] pre the tests, though [16:38] tsdgeos, so no processEvents() loop ;) [16:38] :D [16:53] mterry: hmm, what does "requires android platform bits" mean here? I haven't looked in too much detail at the qtubuntu source, I think you want ChickenCutlass for that, but in general the androidiness should be cross-platform in that we don't actually have it in the archive on any arch :) [16:54] mterry: however, the current qtubuntu code may have arm-specific assumptions in its own right - since IIRC this is one of the components that deals with libhybris === _salem is now known as salem_ [17:10] * ogra_ LOLs about the "how to enable mouse cursor" thread on the ML [17:11] (why didnt i nothice it before ?) [17:17] Anyone familiar with running the unity/phablet autopilot tests? I am getting errors about the autopilot interface not being available despite having autopilot-phablet installed === dandrader is now known as dandrader|lunch [17:29] ogra_: reading your mp, I was wondering whether we should use upstart user sessions instead of the shell script spawning the processes [17:29] (obviously that's raring only) [17:33] lool, right, we should, though i'm not sure about the delays ... for now the most important bit is the use of ubunut-session.d, so i'd still like to see it being merged [17:34] there is another /bin/bash script in that branch that really has a penalty to startup speed, i plan to look into that on the weekend [17:34] (create-online-accounts) [17:41] robru, kenvandine, cyphermox: have any of you messed with running autopilot tests for the phablet unity branch by any chance? [17:43] lool, fixed :) [17:45] mterry, no, sorry [17:48] mterry, i haven't [17:53] ok, anyone who can help with networking (or missing network interface)? [17:54] dun1982, got the module for your device ? [17:58] Well I enabled wifi drivers from the kernel setup, so they should be compiled and deployed along with the rest of the kernel. [17:58] Is there anyway to check if I have the drivers or not installed in the kernel? [17:58] ie. lspci is not available and /etc/network/interfaces has only lo [17:59] cat /proc/net/dev [17:59] thats the most lowlevel way of checking for the device [18:00] (at least the most lowlevel one i know) [18:01] yup went there and the hardlink net does not point anywhere. [18:01] cat /proc/net/dev [18:01] cat: /proc/net/dev: No such file or directory [18:01] mount /proc then :) [18:02] should I mount it inside chroot or from android? [18:02] ah, from android I get in to /proc :D [18:02] right, /proc isnt mounted when you chroot [18:03] so after chrooting, mount proc (i think the release notes say so) [18:03] ah, I go and read the release notes then. [18:03] I did not know that we had one :D [18:04] /topic [18:04] ;) [18:05] ogra_: when you say the scripts are a penalty to startup speed, does it mean you have some boot charts somewhere? just curious as I'm a bit surprized these would make a bit difference [18:06] Btw, the release notes says that you should be in ssh-connection before running the mounting /proc. [18:06] Kind like a egg chicken situation if you ask me... [18:06] lool, nope, only gut feeling [18:07] lool, the PID handling the script is crap btw ... since there are varous scripts that only firs once and start sub processes with different PIDs ... so after all we definitely want upstart [18:07] dun1982, i mean, if I wanted to attach a flash drive or usb drive to my touch device, whether by plugging straight in or cracking the device open and soldering it on, is ubuntu going to recognize the size? Theoretically is it possible, will ubuntu drivers pick it up. Is the limitation in hardware or is it in the OS [18:07] *fire [18:07] dun1982, (re: the requestion I asked yesterday) [18:08] froek: sorry, I have no idea what are you asking :) [18:08] ie. are you asking from me if you can solder a usb-key into your device? [18:08] i'll try a different way [18:08] I want to add storage to my touch device (somehow) [18:08] will the OS recognize the drive to it's fullest [18:09] froek: well there is no such option I am afraid. If you do not know where to solder the device into. [18:09] dun1982, lets say I do know where to solder it.. [18:09] Well then do you know from software level how to use the signals you are getting after the soldering? [18:10] ie. after soldering I am quite sure your device does not pick up the new added hardware. [18:10] Not at least automatically. [18:11] Bit let's say you can replace exising memory cell with bigger one, that _might_ work :) [18:11] dun1982, yeah, replacing existing is kinda where I'm going [18:11] but seriously, I would still suggest to buy a device with sdcard place. [18:11] I'm curious if there would be development required in order to further the drivers software, and not just the hardware which I know I'll be needing to do [18:11] A lot simpler solution, otherwise I could not help you that much. [18:12] no worries.. with SD yes, there is that option - but it's hard to get beyond say, 64GB [18:12] I'm looking at a multiple of that [18:13] I realize I'll need to do some hardware hacking, which is fine by me [18:13] Is the touch version of ubuntu that different than the main Desktop/Server edition? [18:13] eh, okay why would you in right mind go for more than let's say 32gb? [18:14] umm, more of an exercise in fun to go beyond 32 :) [18:14] yeah. it is not the desktop version, but thinned down version of broken software :) [18:14] its bnot broken ... its just young :) === dandrader_ is now known as dandrader [18:14] ha, I see. Is it possible to upgrade to full version using apt? [18:14] froek: no [18:15] ie. for you it might be better to try to port the desktop over the device you want to use :) [18:15] hmm [18:16] But the software and platorm is quite open, so I would imagine that anything is possible. [18:16] But I have no answers for how to do what you want. [18:16] Not a problem, you've helped me already [18:16] And I do not think that there are that many others who want to do what you are trying to do, so you might need to find the answers alone :( [18:16] where would I find the source for the touch specific os? [18:17] i've got the developer preview tools installed [18:17] but is there a main code repo that is what makes up the touch OS? [18:17] Hm... I have no idea where the touch source is available. [18:17] I only know where to find the daily binaries. [18:18] oh, where's that? [18:18] I could probably track it down [18:18] http://cdimage.ubuntu.com/ubuntu-touch-preview/ [18:19] tx [18:20] froek, you want /current [18:21] cheers [18:22] anyone here know who/where the "build" people are for ubuntu-touch images? [18:23] they are little gnomes in automated buildservers [18:23] * froek looks high and low [18:25] froek, what do you want to know [18:26] ogra_, I'd like to pull the source tree and learn how those images are built, maybe even contribute back to the project [18:26] froek, well, see the porting wikipage (channel topic) [18:26] k [18:27] ogra_, thx [18:27] the image is actually two parts ... [18:27] one is the android base which is cross built from a modified CM tree ... [18:27] the other is a typical ubuntu rootfs assembeld from debs [18:27] I see that in the first paragraph of the porting page.. interesting [18:27] interesting [18:28] it's kind of sad that the part everyone usually complains is not "open" enough, android, is currently easier to build from scratch than the "real" linux part [18:28] so you're basically running a baseline android kernel, and then bootstrapping the ubuntu system underneath that [18:29] ubuntu runs in a container [18:29] using libhybris to access the android side and talk to the drivers [18:29] now, that container is not like the current hack to make ubuntu work on android is it? [18:29] the android side is just the most minimal stuff you need to make the drivers work [18:29] its a bit different, but similar [18:30] there is work going on to actually move the android part into a container though [18:30] what was decided for that? [18:30] and use the ubuntu rootfs nativelz on top of the android kernel [18:30] ali1234, research iirc [18:30] so still undecided then? [18:30] with ubuntu in LXC as fallback [18:31] afaik yes, but i'm most likely not up to date [18:31] all i know is what i saw at UDC session [18:32] UDS* [18:33] yeah, i think ChickenCutLass wass researching the container stuff [18:33] i see [18:33] and based on his freedback it will be decided === dandrader is now known as dandrader|afk [19:17] Is ever going to be ubuntu touch for the htc thunderbolt?? === dandrader|afk is now known as dandrader [19:32] YO [19:33] does anyone have the march 3rd build? [19:33] i need it [19:33] anyone? [19:35] it gets auto-deleted if there are more than two newer ones [19:35] anyone with a backup? [19:36] i need it so much right now [19:37] like over 900000000000000kkkkkkkkkkkkkkkkkkk times appreciated if you have a backup [19:37] * ogra_ doesnt [19:38] :( [19:38] darn [19:38] guess i'll need to use http://cdimage.ubuntu.com/ubuntu-touch-preview/quantal/mwc-demo/quantal-preinstalled-phablet-armhf.zip [19:38] why is the 3rd so important ? [19:41] is there a daily snapshot avaible? [19:42] 3rd is the last one known to work on samsung devices [19:42] dues to what exactly ? [19:42] my i9100 works fine with daily dist upgrades here [19:43] which should get me the same [19:43] at least on the phjablet side [19:43] *phablet [19:43] Ive got a samsung galaxy nexus :) [19:43] im installing it at the moment [19:43] the daily ? [19:43] but it hangs for some reason at Pushing /home/ubuntu/Downloads/phablet-flash/95/quantal-preinstalled-phablet-armhf.zip to /sdcard/autodeploy.zip [19:44] ah nevermind :) writing system. [19:45] yeah, its quite slow to transfer [19:45] yep. took almost 6 minutes [19:47] samsung as in the new galaxy devices [19:48] all of the non-S2 or earlier don't like the new builds [19:49] "i killed 3 pigs today working on this" [19:49] does Mobile Data work now_ [19:49] ? [19:49] nope [19:50] dank101, pigs arent enough... zou need a male dragon [19:50] *you [19:50] and 5 dragons [19:50] female and male [19:51] whats the issue you see ? [19:52] blackscreening [19:52] no GUI [19:52] Never loads [19:52] what about SGS? [19:52] any logs [19:52] posted to the mail list about 66666666 times [19:52] my SGS2 works fine here [19:53] not sure about other models ... [19:53] but i know the guy rilling the device image had to jump through some hoops to roll back the mali driver [19:53] *folling [19:53] bah [19:53] *rolling === Lexmazter is now known as anders73X === anders73X is now known as ifyoucanreadthis === Maroc-OS_ is now known as Maroc-OS [20:28] Oh man, I have no idea what to do with the damn image. Now I have included both old Mali driver and wifi drivers, but not screen nor network drivers are available after boot. [20:30] which wifi driver is that ? [20:30] dhd ? [20:34] ath6kl [20:35] # Wifi [20:35] BOARD_WLAN_DEVICE := ath6kl [20:35] WPA_SUPPLICANT_VERSION := VER_0_8_X [20:35] BOARD_WPA_SUPPLICANT_DRIVER := NL80211 [20:35] BOARD_WPA_SUPPLICANT_PRIVATE_LIB := lib_driver_cmd_ath6kl [20:35] WIFI_DRIVER_MODULE_PATH := "/system/lib/modules/ath6kl.ko" [20:35] WIFI_DRIVER_MODULE_NAME := ath6kl [20:35] WIFI_DRIVER_LOADER_DELAY := 1000000 [20:35] BOARD_HAVE_SAMSUNG_WIFI := true [20:35] does /system/lib/modules/ath6kl.ko exist ? [20:37] Do you mean in the image which I have booted or in the out/target/product/p6800 folder? [20:38] in youtr running phone indeed [20:38] There is no modules folder under system :) [20:38] Should I be worried? [20:38] i wonder if your flashing worked then [20:38] Filesystem Size Used Free Blksize [20:38] /mnt/secure 390M 0K 390M 4096 [20:38] /mnt/asec 390M 0K 390M 4096 [20:38] /mnt/obb 390M 0K 390M 4096 [20:39] /data/ubuntu/dev 390M 48K 389.9M 4096 [20:39] /data/ubuntu/run 390M 44K 390M 4096 [20:39] /data/ubuntu/run/shm 390M 76K 389.9M 4096 [20:39] /data/ubuntu/run/lock 5M 0K 5M 4096 [20:39] /data/ubuntu/system 818.9M 168.8M 650.1M 4096 [20:39] /data/ubuntu/vendor 818.9M 168.8M 650.1M 4096 [20:39] /data/ubuntu/data 12.9G 5.8G 7.1G 4096 [20:39] Did that come all the way? [20:39] * ogra_ points to paste.ubuntu.com [20:39] with both hands ... [20:39] http://paste.ubuntu.com/5617633/ [20:40] I need to go for a moment now. [20:40] well, /system should have been filled with your build ... [20:40] and should definitely contain the ath6kl.ko file [20:40] Be back in ~1h or going to sleep anyway :) See u when I'm around next time. [20:41] enjoy [20:41] http://paste.ubuntu.com/5617636/ [20:41] There is my system folder. I think I need to check if I have to upload the system folder by hand there... [20:41] and /system/lib/modules ? [20:42] no, that should al happen from the autodeploy.zip [20:42] you shouodnt need to upload anything beyond that [20:42] http://paste.ubuntu.com/5617639/ [20:43] ath6kl.ko [20:43] there you go [20:43] So what do I do with that [20:43] `? [20:43] so go out of the container to android and do: [20:43] insmod /system/lib/modules/ath6kl.ko [20:44] Ah thx, I'll try that next. [20:44] then cat /proc/net/dev again ... and you should see a wlan0 or so [20:44] should that be automatically done? [20:44] android usually does that from init.$devicename.rc [20:45] for me the line to load the wifi module was missing in the CM tree, i had to add it [20:56] Hi people! [20:57] It's there latest rom for gt-i9300 I wanna tested [20:58] *test it [20:58] mzanetti, alright, I've double-confirmed that autopilot-phablet isn't quite working (or something isn't). I set up a new VM, only did ./build -s and installed autopilot-phablet. Still I get the error about the autopilot interface [20:58] kheeper, check https://wiki.ubuntu.com/Touch/Devices [20:59] mterry: it works here and in jenkins... very likely to be something with your setup [20:59] not out for the i9300 [21:00] I know it's not out for i9300 [21:00] mterry: does it still say "Loading testability driver" when you launch it with -testability? [21:00] just was asking if there is some beta version [21:01] mzanetti, hrm, yes. But then immediately segfaults :-/ [21:02] Maybe that's the real issue. But I don't know why it would. This is a clean setup [21:02] mterry: hmm... and it does not segfault when you start it without -testability? [21:02] mzanetti, it also segfaults then [21:03] no [21:03] seems to be loading libUnity-qml.so [21:03] mterry: then you have probably libunity-core not up to date [21:03] mterry: you need the version from the phablet-team ppa [21:03] mzanetti, the build script is designed to point to a version built from bzr [21:03] mterry: be aware that isntalling that kills old unity for now [21:03] mterry: I haven't ever ran that script so far [21:04] mzanetti, well, I'm running libunity-core from bzr [21:05] from lp:unity/phablet-mods ? [21:05] mterry: ^ [21:06] mzanetti, yes [21:07] mterry: hmm.. that should work I guess. haven't tried that so far. I'm using the libunity-core from the phablet-team/ppa [21:07] mterry: which should be built from that one afaik [21:09] * mterry is tired of trying to get unity/phablet in a working state [21:10] aha... hmm [21:10] Heh, can someone tell me what the heck is wrong... [21:11] I bet I and the autopilot script are both not setting LD_LIBRARY_PATH [21:11] dun1982, well, did you try the things above ? [21:11] http://paste.ubuntu.com/5617704/ [21:11] yeah... [21:11] Check what happened :) [21:12] funny [21:12] yeah, that was my first reaction too... [21:12] Well, if this would have been easy, it would not be called compiling kernel :) [21:13] so next thing ... modinfo /system/lib/modules/ath6kl.ko [21:13] mzanetti, yup, I get farther by manually setting LD_LIBRARY_PATH when running tests (duh). [21:13] check what parameters the module takes [21:13] Man, I start feeling that 15 days is enough when trying to figure out how to get the basic necesities working with the device, so that I could try out the preview version of this os :) [21:14] there might be something like a firmware= parameter or so that you need to point to the right binary blob [21:14] ogra_: http://paste.ubuntu.com/5617707/ [21:15] ah, so it has hardcoded paths [21:15] ok, where in the build process configuration is that enabled? [21:15] athwlan.bin is what you want to find i would guess [21:16] likely in the extracting of the binary blobs [21:16] ok, did this, waiting now results: root@android:/ # find / -name athwlan.bin [21:16] ok, there is no such file in anywhere in the booted image. [21:17] some drivers take a firmware= parameter ... others just have hardcoded paths they try to load from ... and there are some where you have to echo the firmware file into a sysfs path [21:17] I'll check if any of the files are available in the image I flashed to my system. [21:17] well, look in your device tree on the PC [21:18] it could even be that they are included into the .ko file at compile time, though that would be a GPL violation i think [21:19] $ find . -name athwlan.bin [21:19] £ [21:19] € [21:19] So no athwlan.bin anywhere in the whole repository. [21:19] I did run that command in the repository root. [21:19] but you extracted the binary blobs when setting it up ? [21:20] Yeah I did. [21:20] And I'm running cm-10.1 build in my device. [21:20] well, flash cm 10.1 and look exactly what that does then [21:20] my kernel and device threes are based on the original cm-10.1 repositories and I have asked help also from the maintainers of those repos. [21:21] if wlan works you should be able to find out how it does that by liooking at all the things we went through today [21:21] ok, let's see what that turns out then. [21:21] thx anyway. You are one patient teacher :) [21:21] looking at init.rc for an insmod line ... looking for the blobs etc etc [21:29] ogra_: http://paste.ubuntu.com/5617731/ [21:29] This is from cm-10.1 and I have connected it actually to my wlan as well. [21:30] How ever ifconfig or iwconfig do not give any devices... [21:30] But... wlan0: 52194043 34276 0 0 0 0 0 0 1405594 19763 0 0 0 0 0 0 [21:33] well, look for the binary blobs .. have a look at init.rc [21:33] i.e. [21:33] grep insmod /init*.rc [21:33] Yeah, I'm just looking into there. [21:33] Hrmm... now I'm getting the Music page instead of the Home page by default [21:33] to find all insmod lines in all init.rc files [21:34] root@android:/ # grep -ir insmod /init*.rc [21:34] 1|root@android:/ # [21:35] intresting [21:35] say to least... [21:35] does ldmos show any modules loaded ? [21:35] err [21:35] lsmod [21:35] root@android:/ # lsmod [21:35] ath6kl 204531 0 - Live 0x00000000 [21:35] well, something loaded it [21:35] let's see what happens if I grep that aht6kl.ko :) [21:36] or just ath6kl [21:36] its a binary ... grep wont help much [21:36] "strings" would [21:36] but i doubt android ships that [21:37] I have the ath6kl source if that is any help [21:37] It needed to be included in kernel build btw. [21:37] There was no blobs for it. So it is actually build along the kernel. [21:38] https://github.com/dun1982/android_hardware_atheros_wlan [21:38] I had to include it to my manifest.xml because my images were failing without it. [21:39] thats only wpa_supplicant stuff [21:39] not kernel [21:39] But isnt the wpa_supplicant compiled in kernel build. [21:40] This is the full kernel source: https://github.com/dun1982/android_kernel_samsung_p6800/tree/v0.3 [21:40] That is the source that was compiled for my current test image. [21:41] The v0.3 where the only difference where the added ath6kl drivers with that wpa-supplicant stuff in the manifest.xml. [21:42] https://github.com/dun1982/android_kernel_samsung_p6800/tree/v0.3/drivers/net/wireless/ath/ath6kl [21:42] thats your driver [21:47] did you search for the binary blobs on CM already ? [21:49] nope [21:55] where does those binary blobs go which the extract-files.sh load? [21:57] no idea, its usually described in the proprietary*.txt files [21:58] hm... if my find command did actually find the file, they are deposit to either out/.. or nowehere. [21:59] http://paste.ubuntu.com/5617792/ [21:59] no, i mean the binary blobs ... and i mean on your running CM === _inf_ is now known as _inf [22:09] http://forum.xda-developers.com/showthread.php?t=2191235&page=2 [22:16] dank101, and it just works flawless with the mwc-demo ? [22:16] (or dailies from before the 3rd) [22:16] i think [22:16] you think or you know ? [22:16] think [22:17] worked on the VZW S3 [22:17] then test with the mwc-demo to prove your theory first :) [22:17] ok [22:18] ADB works on all of them GUI only works on the dailies before [22:18] right, so your android side is at least basically functional === Ursinha-afk is now known as Ursinha === Lloir is now known as Lloir|zZzZzZz === Ursinha-afk is now known as Ursinha [23:14] hello. it possible install ubuntu on a tablet that does not has android, any link that can help me ? thanks [23:17] impossible [23:17] unless android is ported [23:17] 99999999999999999999999999999999999999% impossible === salem_ is now known as _salem [23:29] Anything interesting today? [23:29] I couldn't catch any of the SDK stuff === Ursinha-afk is now known as Ursinha