=== Ursinha is now known as Ursinha-afk === Ursinha-afk is now known as Ursinha === XenGi is now known as XenGi_ [00:32] sergiusens: rsalveti: heh, phablet-tools, I didn't wake up about it, no point now what there has already been a release in PPA... === Ursinha is now known as Ursinha-afk === Ursinha-afk is now known as Ursinha === Ursinha is now known as Ursinha-afk === Ursinha-afk is now known as Ursinha [01:29] sergiusens: https://code.launchpad.net/~phablet-team/phablet-extras/ofono-release-112phablet1/+merge/158261 [02:18] rsalveti, are you here? [02:19] does anyone who knows about ofonod and ril can help me? [02:25] anyone? === Ursinha is now known as Ursinha-afk [02:41] hello [02:42] I flashed ubuntu-touch to my Sony Arc from http://forum.xda-developers.com/showthread.php?t=2226406 [02:42] It's boot but performance is low [02:44] I checked by "top" and find that, hud-service always around 54%, rsyslogd is 44% [02:45] Could anyone help me? I think if hud-service and rsyslogd can be disabled or reduce the cpu usage, ubuntu-touch may be run good in my phone [02:52] anyone on ril or ofono that can help me? === Ursinha-afk is now known as Ursinha === freeflyi1g is now known as freeflying === Ursinha is now known as Ursinha-afk [03:28] Ursinha-afk, are you there? === gxxxyyy is now known as gregk === gregk is now known as gregk8 === gregk8 is now known as gregk === Ursinha-afk is now known as Ursinha === Ursinha is now known as Ursinha-afk === Ursinha-afk is now known as Ursinha === chriadam|away is now known as chriadam === Ursinha is now known as Ursinha-afk === Ursinha-afk is now known as Ursinha === Ursinha is now known as Ursinha-afk === Ursinha-afk is now known as Ursinha === Ursinha is now known as Ursinha-afk === msorvig_ is now known as msorvig === Ursinha-afk is now known as Ursinha === Ursinha is now known as Ursinha-afk === Ursinha-afk is now known as Ursinha === Ursinha is now known as Ursinha-afk === Ursinha-afk is now known as Ursinha [06:46] good morning === Ursinha is now known as Ursinha-afk [07:08] good morning === Ursinha-afk is now known as Ursinha === Ursinha is now known as Ursinha-afk === Ursinha-afk is now known as Ursinha [08:00] hi [08:01] nobody === Ursinha is now known as Ursinha-afk === Ursinha-afk is now known as Ursinha === Ursinha is now known as Ursinha-afk === Ursinha-afk is now known as Ursinha === ckpringle_ is now known as ckpringle === Ursinha is now known as Ursinha-afk [08:47] I'm curious, could someone explain how the graphics stack works right now ( the X bits ) on ubuntu touch? [08:47] more specifically to the N10 [08:49] ogra_: do you have any idea ^ ? [08:49] no X bits involved :) [08:49] dpm: ping [08:50] hey oSoMoN [08:50] in the android layer you have SurfaceFlinger which the Ui attaches to from the ubuntu container via libhybris [08:50] hola dpm [08:50] buenas :) [08:50] that part will soon be replaced by Mir [08:50] dpm: do you know if jenkins/autolanding is enabled for the core app template branch? [08:51] dpm: I have https://code.launchpad.net/~osomon/ubuntu-phone-commons/appTemplate-packaging-fixes/+merge/157077 which is approved, but hasn’t landed yet [08:51] shadeslayer, see the Mir spec pon the ubuntu wiki, it has an overview of "today" and "tomorrow" [08:51] s/pon/on/ [08:52] ogra_: I see, and what if for Kubuntu we want to use X + kwin_gles [08:52] since kwin won't have Mir integration [08:52] ( atleast that's the situation right now ) [08:52] you will be able to use X as you do now [08:52] but graphics acceleration? [08:52] oSoMoN, I think it's only enabled for the core apps. If you think it makes sense for it to autoland, I think we can just ping mmrazik (who doesn't seem to be online right now) and add it to jenkins. For now, I could merge in your change manually [08:52] we dont plan to drop it from the archive :) [08:53] once Mir enters the desktop there will also be XMir [08:53] haha, no, I mean, lets say I install ubuntu-touch and then install Kubuntu ontop of that [08:53] ah, that wont work (doesnt work with gnome or other stuff either) [08:53] dpm: not sure autolanding is really needed for this branch, but if you can merge mine I’d appreciate :) [08:53] shadeslayer, you would need to port kwin to QtMir [08:54] right, so my question is, is there a way to get the drivers from https://developers.google.com/android/nexus/drivers#manta, plop them somewhere and make it work? [08:54] or just live without Kwin ... the apps should just work [08:54] ogra_: uh, Martin's position is pretty clear on Mir atm [08:54] btter ask that in #ubuntu-mir [08:55] yoou cant just use the android drivers without their context i guess ... you will need some android env around it [08:56] (the kernel and HAL layer) [08:56] I see === Ursinha-afk is now known as Ursinha [08:58] in any case the guys in #ubuntu-mir are far more qualified than i am, i'm mostly guessing based on what i know today ... [08:58] oSoMoN, ok, merged :) [08:58] they can surely go more into detail [08:58] okay will ask [08:58] dpm: awesome, thanks! [09:01] possibly the mer people have also figured out something === Ursinha is now known as Ursinha-afk [09:10] jhodapp: rsalveti: when you will have a moment later today, I'd like to chat about platform-api & ubuntu-sensors and testing. I did run the test_* apps from the platform-api for the hybris implementation and they do work. Are those being run in jenkins setup somewhere. Similarly for qtsensors, there are no tests at the moment - but it's a plugin for qt framework. [09:11] should it like have (1) api-compat unit tests against qt framework (2) sensors regulartly tested (autopilot or manually or some funky integration tests "e.g. change brightness via api & check if kernel reports a delta change) [09:13] xnox, i think that stuff isnt finalized yet [09:13] I mean if either of platform-api / sensors get broken and "slip-in" it will be detected quite quickly. But i'd like some automated safety net against braking phablet images due to bugs in either of those. [09:14] ogra_: ok. Then I'm happy to start daily landing those - tbh. [09:14] ogra_: cause we will have autopkgtests to check buildability for example and I will push for api-compat monitoring. [09:14] didrocks: ^^^^^^ any other thoughts. [09:15] dont quote me on that though ... but i think the majority of sensores isnt even working afaik [09:15] ogra_: i've tried raring chroot and reverted back to quantal/raring frankenstein, because I was not getting wifi on the raring chroot. [09:15] ogra_: i twiddle the brightness slider with my finger in the indicator and it totally works on nexus 7 =) [09:16] xnox: ok, do you mind coordinate on it with mterry? as I think some of his work is dep on yours [09:16] yeah, no network is a known bug [09:16] ogra_: ok. [09:16] xnox: sad that we can't have them being running before or as part of the daily release [09:16] should be fixed this week though according to sergio [09:16] xnox: meaning, we are breaking something and then fixing [09:16] rather than the other way around :) [09:16] loicm: could you sponsor lp:ubuntu-ui-toolkit (0.1.41) into raring? I merged your raring specific changes so it should be fit for archive upload as well (with the general FFe for mobile related stuff) [09:16] xnox: please put all details in the spreadsheet so that we can have a summary [09:17] didrocks: true. but we are a bit in a chicken & egg situation: cannot test this stack (e.g. raring/s daily image flashed to a tablet), since we don't have all the pieces built/bootstrapped/image generated. [09:18] xnox: yeah, just mark that down and check with mterry to have the stack daily landing now [09:18] xnox: you have the bootstrap commits and everything? [09:18] once we have first image, we can start doing things similar to daily-iso-testing-desktop. [09:18] didrocks: will fill the details on the spreadsheet and commit bootstrap marker. === Ursinha-afk is now known as Ursinha [09:18] otherwise sensors are fine. and platform-api is landing into ppa already. [09:19] ogra_, Hi, as I remember you know how to detect version of build image? [09:19] xnox: great! :) [09:19] P3T3: .disk/info or some hidden file?! [09:19] P3T3, somewhere in /system/etc ... [09:19] oh..... [09:19] there is a stamp file [09:20] or was it /etc ? ... /me forgot [09:21] xnox, that will likely be there once we do actual cdimage builds :) [09:21] in S or so [09:21] OK there is /etc/buildstamp [09:21] right [09:21] didrocks: hm.... your comments seemed to be gone from the spreadsheet? or they got marked "resolved" [09:22] xnox: urgh? [09:22] didrocks: nevermind found them. [09:22] yep ;) [09:22] I have an image from popey with acubens Fri, 01 Mar 2013 05:23:11 +0000 [09:23] This image is later then MWC demo and it boots in UI, is this image from daily builds? [09:23] yes P3T3 [09:23] P3T3: march 1st is the datestamp on it [09:23] I know, but do not know which date starts daily images :-) [09:25] P3T3: how will that help? [09:25] i think the first public build was on feb 19 [09:25] or thereabouts [09:26] popey, latest daily builds fails to boot in GUI, I am searching what changed and when [09:27] sure, but you said the one you have works? [09:27] so surely something broke _after_ the image you have, not before [09:27] this one is from you and it works [09:28] right, and that's from march 1st, so you want to find out what broke after? [09:28] thats right! [09:28] we can do a bisect ☻ I can upload one from half way between then and now [09:29] I think enough will be date 10th March [09:29] ok, thats build 17, want that one? [09:30] if you can :-D [09:30] sure [09:31] P3T3: it's uploading to the same place, will be there later [09:31] popey, Thank you, I will wait [09:33] np [09:35] otherwise, I think this image from March 1st is a bit faster then MWC one [09:35] thats possible [09:45] ogra_: I'm reading up on the xf86-video-armsoc , which seems to be a driver for the Mali T6xx series [09:46] shadeslayer, well, yeah, but it has its issues with composite (i'm using it right now) [09:47] oh [09:47] i thought you were after running kde on top of ubuntu touch [09:47] hm, nope, I was thinking of running X -> Run KDE [09:48] ah [09:48] well you asked about the graphics stack of ubuntu touch initially [09:48] so i kind of assumed that [09:48] yeah, I was investigating how it works on ubuntu touch :) [09:48] I think I understand now [09:48] k [09:49] armsoc will work indeed [09:49] yay, but with issues, that's fine [09:49] but has its issues and you need to hack around in the ldconfig setup [09:49] atleast there's something I can start off with [09:49] oh? [09:49] (to prevent mesa from taking over) [09:49] ah [09:50] you want mesa for GL stuff but not for GLES [09:50] roger [09:50] and the driver doesnt have a concept fro this (since the GLES driver is completely unlicensed, so cant be distributed in packages) [09:51] fun, do you have your setup documented somewhere? or can you throw it up in a pastebin real quick? [09:52] the N10 has very bad button placement :( [09:53] so awkward to press the buttons at the top [09:53] shadeslayer: all android devices have bad button placement ☻ [09:53] didrocks: lp:qtubuntu-sensors has the bootstrap node, comments added on the spreadsheet, bugs filed against the project and linked to the blueprint to create WI. [09:53] popey: haha [09:53] didrocks: please "add" qtubuntu-sensors to start landing into the ppa. [09:54] i just removed the /etc/ld.so.conf.d/arm-linux-gnueabihf_GLES.conf link, put the libGLES stuff for mali into /usr/lib and run ldconfig here [09:55] xnox: does it have autolanding? (upstream merge) [09:55] xnox: I don't find it in the current stack [09:55] ogra_: 'here' ? [09:55] on my chromebook [09:55] ah okay [09:55] (same HW) === Ursinha is now known as Ursinha-afk [09:57] P3T3: http://people.canonical.com/~alan/phablet/17/ finished [09:58] ogra_: and regarding the libGLES stuff for mali, did you use the libGLES so that google provides or the driver from the archive? [09:58] there is no driver from the archive [09:58] the driver I downloaded from google has a libGLES_mali.so [09:58] as i said, its not distributable [09:59] uh [09:59] i copied the files from chromeos [09:59] http://pad.lv/u/xf86-video-armsoc [09:59] didrocks: hmm... explain. It was part of the blueprint tasks for cleanup & landing. And this: https://code.launchpad.net/~mterry/qtubuntu-sensors/bootstrap/+merge/157399 [09:59] did get merged. [09:59] that doesnt contain any mali libs [09:59] its only the driver [09:59] ah I see [09:59] xnox: yeah, interesting, I don't find the configuration of the CI jobs, not sure how they dealt with it, but looks good, I'm adding it to the daily release stack [10:00] ack. [10:00] didrocks: not sure what to do with this "changelog-entry-only-merge" https://code.launchpad.net/~jhodapp/qtubuntu-sensors/qtubuntu-sensors_release/+merge/152181 [10:01] xnox: seems they did a "release" the old way (they change the changelog) for phablet [10:01] xnox: you can get it merged if you want, would be more clear if the top changelog is the one with the boostrap though [10:01] (but not 100% necessary) [10:02] ogra_: how's ubuntu on the Chromebook though? [10:02] shadeslayer, fine apart from the composite issue ... unity doesnt catch that and runs (even thinks it is fully supported) but it slow as hell [10:02] so i have to resort to other WMs [10:03] waiting for Mir desktop support :) [10:03] heh :) === Ursinha-afk is now known as Ursinha === edulix2 is now known as Edulix === Edulix is now known as edulix [10:04] I can't wait to flash this literally-brand-new-N10 [10:04] alternatively hoping that someone fixes it in early S for arm desktops :) [10:05] xnox: added to the daily FYI [10:05] * ogra_ loves unity (like manu others) on arm desktops ... so it would be nice if that cound be fixed at some point [10:05] s/manu/many/ [10:06] didrocks: cool, thanks. [10:06] thanks to you :) [10:06] ogra_: just to make sure I understood this correctly, the armsoc is the driver, that's opensource and licensable, OTOH samsung ships it's own libGLES that cannot be packaged, correct? [10:06] right [10:06] okay [10:06] and armsoc uses the libGLES for acceleration if it is available [10:06] right [10:11] popey, downloaded and now in flashing process. I will inform you about result === Ursinha is now known as Ursinha-afk [10:18] P3T3: great [10:20] popey, fails, have black screen [10:21] bummer [10:21] can we try other build ? [10:21] I have Mar 1st working and MAr 11th non working [10:22] first one is build #8 and second one is #17 === Ursinha-afk is now known as Ursinha === chriadam is now known as chriadam|away === Ursinha is now known as Ursinha-afk [10:40] P3T3: sure [10:40] P3T3: i have 9, 10, 11, 13, 16 === SkavenXXI-[OFF] is now known as SkavenXXI [10:42] popey, 13 as it is happy number in China [10:42] P3T3, but it isn't in Western countries (I'm from China) [10:43] my house is 13. but the stupid builders were superstitious so they labelled it 12A ☹ [10:43] My house number is 13 too === Ursinha-afk is now known as Ursinha [10:44] P3T3: uploading, same place [10:45] popey, and you can delete 17 if you no longer need it [10:45] thanks === MacSlow is now known as MacSlow|lunch [11:05] Hi all. Does 3g data work on current build? [11:06] nope [11:06] :/ So still have to wait. Thanks. [11:07] but it is actively being worked on [11:15] ogra_: the phablet-armhf.zip is basically just a debootstrapped ubuntu + packages right? === Ursinha is now known as Ursinha-afk [11:34] popey, #13 is black too. So remains 9,10,11. Can you give me 10 please? 13 is to delete [11:35] bzoltan: is there any reason the 1.41 toolkit release is not in raring? [11:35] 0.1.41 [11:35] tsdgeos: we do not (can not ) release straight to Raring [11:35] tsdgeos: so there is a delay between our release and the distro release [11:35] ok [11:36] tsdgeos:I suggest to use the SDK Release PPA if you need the latest Toolkit [11:36] that's what i'm doing [11:36] but using ppa's makes me sick :D [11:39] hello all [11:44] hi [11:44] shadeslayer, well, it is ubuntu-minimal + android integration + ubuntu touch UI [11:45] right, I'm just trying to create my own zip [11:45] seems like there are some other files in there too META-INF/com/google/android/update-binary [11:45] yeah, android stuff [11:47] where does that come from? [11:48] aha [11:48] phablet-dev-bootstrap gens the zip [11:49] xnox: we can chat about that whenever you want [11:52] jhodapp: well, I think between myself, ogra & didrocks we agreed to start autolanding qtubuntu-sensors. We just need to come up with testing strategy for it. [11:53] * ogra_ is out for ~2h [11:53] xnox: indeed [11:53] jhodapp: are sensors api exercised via autopilot tests somehow at the moment? or since they are - well - sensors they need manual testing? [11:54] P3T3: ok [11:54] xnox: manual testing, although we should be able to add some simple unit tests to some of those functions [11:54] I have filed http://pad.lv/1167825 http://pad.lv/1167818 for pure api/abi stability monitoring. [11:54] Launchpad bug 1167825 in qtubuntu-sensors "needs reverse-dependency testing" [Undecided,New] [11:54] Launchpad bug 1167818 in qtubuntu-sensors "needs api stability tests" [Undecided,New] [11:54] xnox: though I'm sure someone could get creative in simulating the sensors [11:56] jhodapp: well, we don't want to mock the sensors, cause that's not the point. When I ask to dim the screen - it should dim for real, cause it's a bit pointless to say that well the mock sensors does dim... but the actual tablet regresses and doesn't change brightness. [11:57] I wonder if Certification team has something in checkbox to test that already - e.g. testing the britness keys on the desktop. But instead of manually pressing buttons, we can go via qtubuntu-sensors api to up/down brightness. [11:57] jhodapp: what type of unit tests were you thinking? to exercise / checks all functions. Is there like a standard qtsensors examples which we can be compiling/running atainst qtubuntu-sensors plugin? [11:58] xnox: right but that's a functional test...I'm talking more unit testing [11:59] xnox: really basic ones, qtubuntu-sensors really doesn't do a whole lot...it's just very simple glue between the AAL layer and a Qt/QML app [12:00] * Namidairo sniffs qtubuntu-sensors [12:00] jhodapp: ok. well if you have something in mind that could be done, please open a bug against sensors or add a WI to the landing raring touch blueprint. [12:00] jhodapp: would you be willing to write those? [12:01] xnox: I don't have anything specific in mind atm, but if I think of something I will certainly add it [12:01] xnox: a bug/blueprint entry [12:02] jhodapp: ok. I'll poke the code again once we get the raring/s images. [12:02] xnox: also know that I'm no longer the person actively working on that code [12:03] popey, if 10 will work then I will need 11, otherwise 9. But first I will need to test it. Will inform you [12:03] jhodapp: hmm.... who became "goto person" about it? [12:04] P3T3: np [12:04] xnox: I'm not sure offhand, but I can find out for you if you like [12:04] jhodapp: please do. I'm not familiar with the upstream teams hacking on this stack. I'm from ubuntu engineering trying to get it to land into archive =) [12:05] xnox: ok, np then === greyback is now known as greyback|lunch === MacSlow|lunch is now known as MacSlow [12:20] popey, downloading 10, thanks === Ursinha-afk is now known as Ursinha === LarrySteeze is now known as LarrySteeze|Away === _salem is now known as salem_ === Ursinha is now known as Ursinha-afk === greyback|lunch is now known as greyback === oreneeshy_ is now known as oreneeshy === Ursinha-afk is now known as Ursinha [13:07] popey, #10 fails. Remains #9, can you upload it? [13:07] sure [13:09] popey, 10 & 13 you can delete [13:10] done, thanks [13:11] me too === bzoltan1 is now known as bzoltan [13:15] Hi, may I ask if UT caters for phones with hardware buttons ? [13:17] the UI is designed to not require hardware buttons, but I think some ports have mapped them to certain functions [13:18] thank you pmcgowan I shall try to search for said 'ports' [13:18] id0_stupid_user, what phone do you use? [13:19] Try to find yours at https://wiki.ubuntu.com/Touch/Devices [13:19] I have Nokia N8 [13:19] is it possible to install ubuntu on NOkia n8 [13:19] Upender, whoa I don't think so. Someone needs to port Android to Nokia N8 first:P [13:20] smartboyhw: I have an old android tablet that I pushed generic ubuntu LTS armhf to. [13:20] okies..actully i am not happy with nokia belle any more [13:21] hence looking to chagne but no issue will check for any other update or possiblity in furthre [13:21] future [13:21] I've got the hardware buttons working via a C script that I wrote... Obviously not having X I can't use xevents [13:21] Upender: cool ... My wife has an N8 - I need to repurpose that too ;) [13:23] N8 is symbian, good luck rooting it [13:23] :) [13:23] hmm [13:23] Upender: if you just hate belle there is a secret way to downgrade to S^3 === Ursinha is now known as Ursinha-afk [13:30] Good luck ladies and gentlemen [13:35] uh hi [13:36] when running adb root I get : adbd cannot run as root in production builds [13:36] causing phablet-flash to fail [13:36] any ideas? [13:37] P3T3: 9 is done [13:39] popey, thanks, going to test it [13:45] mmm [13:45] I'll have to root it? === Ursinha-afk is now known as Ursinha === dandrader is now known as dandrader|afk [13:49] xnox: you'll want to talk to ChickenCutlass about the sensors now, he's leading that charge [13:49] jhodapp ok, thanks. [13:50] xnox: np [13:51] popey, #9 works, it's buildstamp = arneb Sat, 02 Mar 2013 05:02:52 +0000 , manhattan-quantal-armhf-20130302-1 [13:53] P3T3: good luck finding what changed ⍨ [13:55] popey, is there any log with changes? === ckpringle_ is now known as ckpringle === Ursinha is now known as Ursinha-afk === dandrader|afk is now known as dandrader === Ursinha-afk is now known as Ursinha === t1mp_ is now known as t1mp [14:23] hey guys in the ubuntu touch component showcase, I see a tab for icons. I also read the demo code. However where does one find the entire list of icons? [14:24] xnox: ogra_: indeed, the sensors are not "done" at this moment, and I believe the apis will probably change with the new desired implementation at the platform-api [14:24] I see them using call-start, computer-symbolic etc..but I am unaware as to what other icons are available [14:24] ChickenCutlass is the owner of sensors in general, so he might know more [14:24] xnox: but we have the infra available already to run test apps at a real device already [14:24] kaleo: see nik90's question? Do we have a full list? [14:25] so we just need the proper plumbing [14:25] * ogra_ looks for his wrenches [14:25] rsalveti: ack. What about the test_* apps in platform-api? same story? [14:26] xnox: yup [14:26] they were just quick test apps to make sure hybris wasn't failing while doing the development mostly [14:26] as we're redesigning platform-api, I also believe it'll change quite a bit [14:31] popey: nik90: sorry there is no list yet [14:32] popey: nik90: moreover the list is different on the desktop and on touch right now [14:32] popey: nik90: we use ubuntu-mono-dark on the desktop and ubuntu-mobile on ubuntu touch [14:32] kaleo: oh ok ... then I'll wait until we get the icons officially for the touch [14:32] dpkg -L ubuntu-mono will give you the desktop ones [14:32] dpkg -L ubuntu-mobile will give you the touch ones [14:32] nik90: just hang on :) [14:34] kaleo: where do I find the ubuntu-mobile package? === francisco is now known as Guest35830 [14:47] intetesting [14:47] our calculator is mathematically correct [14:47] which means people will complain :D [14:48] well, now i did 12 * 8 and got 9696 [14:48] forget that mathematically correct part :D [14:49] tsdgeos: do you mean in the ubuntu-calculator-app? [14:49] nik90: whatever is shipped in the phone [14:49] i guess yes [14:52] tsdgeos: just tried it now...wierd [15:02] ogra_: once you're back, I'd like to talk :) === dandrader is now known as dandrader|lunch [15:02] nik90: it's part of the ubuntu-themes source package [15:03] nik90: which is or will be soon in raring [15:03] nik90: otherwise I think there is a PPA http://ppa.launchpad.net/phablet-team/desktop-deps/ubuntu/ [15:06] shadeslayer, i'm here [15:14] sergiusens, rsalveti, ogra_: so client-1303-sponsoring-community-touch-builds is for "1303" obviously - I'm just wondering if with the next UDS happening at 14-16 May 2013 it'd be something we should discuss there? [15:14] or do we expect many bits already having fallen into place by then? [15:14] i hope the android builds will happen on cdimage by then [15:14] I use my tablet for web browsing and playing videos, will the preview serve my purposes yet? I'm a long term ubuntu user and wouldn't mind some bugs [15:15] kvarley, both technically work ... but the browser doesnt know about tabs yet for example ... so its not very comfortable as an enduser [15:16] Ok, well my setup is basically TV on my tablet and IM on my phone so I guess I can browse on my phone too [15:17] kvarley: depends on the device too, videos play on my Nexus 7, but sound doesn't [15:17] but a couple weeks ago, videos didn't play [15:18] so, progress! [15:18] ogra_: hey, so, I'm curious, the chroot that phablet-flash flashes, that contains android stuff and ubuntu chroot correct? [15:18] mhall119, just enable subtitles :) [15:18] ogra_: what if I don't want the android tools and just want a minimal ubuntu chroot ( debootstrap + ssh server ) [15:18] shadeslayer, it contains the ubuntu side of the platform api and libhybris [15:19] shadeslayer, you can do that but wont be able to use any graphical stuff i belive [15:19] oh, not even by using the methods we talked earlier? [15:19] you will need to talk to SurfaceFlinger soemhow [15:19] (the android display server) [15:20] shadeslayer, wait for Mir:P [15:20] no no, what if I don't want surfaceflinger and just want to use X? [15:20] or build your android in a way that it doesnt fire that up [15:20] ogra_: if I wanted to read a movie, I'd buy the book :P [15:20] smartboyhw: KWin upstream is not going to accept Mir integration :P [15:20] smartboyhw: Martin made that very clear [15:20] shadeslayer, I know (LOL) so that:P :P [15:20] mhall119, audiobooks ftw :) [15:21] okay, I thought you were serious ;) [15:21] (with subtitles indeed) [15:21] ogra_: unless you have to audio.... [15:21] do audio books have subtitles? [15:21] rsalveti, do you know something about compilation parameters of daily builds? [15:21] shadeslayer, I did watch the whole conversation. [15:21] lol, dunno [15:22] P3T3: sure [15:22] ogra_: recall our conversation earlier today, can't I boot minimal rootfs, put more things that I want ontop of it and then use the libGLES from Samsung to get HW acceleration? ( No Android things at all on the system ) [15:22] dholbach: I believe we should be done with it before next uds, but I'd also vote to have some sort of community porters checkpoint [15:22] rsalveti, ok great [15:22] to see what we could do to improve the porting experience and such [15:22] shadeslayer, you should be able to ... try it :) [15:22] sounds great [15:23] right, now my actual question :D [15:23] oh, you did hide it behind smalltalk ! [15:23] rsalveti, today I have discovered that build from March 3rd causes my problem which leads in black screen [15:23] how do I switch out the zips? :P [15:23] dunno, why do you want zips ? [15:23] because I don't quite understand how the booting works here [15:23] I need to know something about changes between MArch 2nd and 3rd [15:23] oh [15:23] P3T3: hm, we didn't change any parameters back then [15:23] P3T3: which hardware? [15:24] HP Touchpad [15:24] so I can flash quantal-preinstalled-boot-armel+manta.img and then flash a rootfs to the /data ? [15:24] and it will all just work? [15:24] shadeslayer, have a look at the nexus7 codepath of ac100-tarball-installer ... thats what we use for the n7 desktop images ... should workj similar for the n10 [15:24] you want a boot.img and put in an initrd that contains the ac100-tarball-installer bits [15:25] I see [15:25] then you can just use a rootfs tarball you prepare [15:25] might need some hacking but essentially thats the process i would use [15:25] okay, thanks :) [15:25] P3T3: I wonder if that is an incompatibility of the platform-api and the compat layer at hybris [15:25] I'll try and figure out how the AC100 does things [15:26] it might be related with the ubuntu rootfs you're using as well [15:26] rsalveti, shall I make it part of the community track? [15:26] shadeslayer, better look at the nexus7 codepath in there ... thats way simpler [15:26] the ac100 has a ton of checks etc that you dont really want [15:27] dholbach: yeah [15:27] sweet, on it [15:27] rsalveti, my hybris and platform-api are newest one [15:27] P3T3: weird, have the logs from logcat? [15:27] would be nice to compare the ones from the working and not working images as well [15:28] rsalveti, P3T3 is Petr Bláha on the ML btw :) [15:28] ogra_: thanks! [15:28] :-) [15:28] http://pastebin.com/ir3Gdk1K [15:28] from the black screen thread [15:28] http://pastebin.com/ir3Gdk1K [15:28] :) [15:28] I am too slow [15:29] it dies at ubuntuappmanager, I think [15:30] sergiusens, rsalveti, ogra_: https://blueprints.launchpad.net/ubuntu/+spec/community-1305-touch-porting [15:30] ogra_, rsalveti true, I am Petr Blaha :-D [15:30] right, let me try to better find what changed between such builds [15:30] dholbach: great, thanks [15:31] ah, already subscribed [15:34] dholbach: didn't we have a blueprint for this? [15:34] sergiusens: this is for uds [15:34] sergiusens, one for 1303 [15:34] dholbach: ah... I'm still confused by the milestones :-) [15:34] :) [15:35] dholbach: that last one is for the next vUDS, right? [15:35] http://www.reddit.com/r/Ubuntu/comments/1c51ci/building_an_ubuntu_sdk_app_rev_56/ could use some upvoting love [15:35] yes [15:35] dholbach: also, porting is going to simplified a bit since we got breakfast to work and I plan on removing some stuff from the build :-) [15:35] :-D [15:35] awesome [15:36] * sergiusens doesn't know what he's missing but doesn't use reddit === jhodapp is now known as jhodapp|lunch [15:38] rsalveti, there is a logcat from working build = http://pastebin.com/3dRghCH0 [15:38] dpm: do you want to add my blog series to http://developer.ubuntu.com/resources/app-developer-cookbook/mobile/ ? [15:38] or some other category of the cookbook? [15:39] P3T3: can you also get me your /system/quantal-ubuntu_stamp ? [15:39] but guess it'll be dev, nevermind [15:39] mhall119, sure, that sounds good. Feel free to add them, I think that's the right category. [15:39] ok [15:41] rsalveti, at working I have only "dev" inside [15:41] P3T3: open the phablet.*.zip from that image and check the tarfile name [15:43] sergiusens: manifests from march 1,2,3,4 are the same [15:43] working build is : manhattan-quantal-armhf-tar-20130302-1.tar.gz [15:43] weird [15:43] and non working is manhattan-quantal-armhf-tar-20130303-1.tar.gz [15:44] diff -Naur manhattan-quantal-armhf-tar-20130302-1.manifest.appcache manhattan-quantal-armhf-tar-20130303-1.manifest.appcache [15:44] nothing [15:44] anyway, need to grab food, brb [15:44] and what about hardfloat compilation flag? I have this one with software backward compatibility [15:45] dpm: http://developer.ubuntu.com/resources/app-developer-cookbook/mobile/ look good to you? [15:45] rsalveti, hmm, did we add some hardcoded NEON stuff perhaps ? [15:45] remember thats a tegra2 [15:46] mhall119, looks good. Could you add the currency converter tutorial while you're at it? I see I forgot to add it. [15:46] ogra_, but I have same problem wirh HP Touchpad [15:46] dpm: sure [15:47] HP has Snapdragon with neon, AFAIK [15:47] ah, k [15:47] ogra_: There is no change fro manhattan-quantal-armhf-tar-20130302-1 to manhattan-quantal-armhf-tar-20130303-1 ... just reconfirming what rsalveti said [15:47] yeah [15:48] dpm: done [15:48] cool, thanks [15:49] dpm: any other articles you know about that we can link in there? [15:52] mhall119, the one I can think of is Rick's blog post, there might be some others -> http://theravingrick.blogspot.com.es/2013/03/sweet-ubuntu-device-qtcreator.html [15:52] he had a few of them, actually [15:54] rsalveti: ogra_ do you recall where that NVIDIA_HACK was? [15:55] NVIDIA_HACK ? [15:55] whats that > [15:55] ? [15:56] rickspencer3_: do you have any problem with us linking to your blog from http://developer.ubuntu.com/resources/app-developer-cookbook/mobile/ ? [15:56] mhall119: what about asking Stuart Langridge for tutorials...I remember him creating many apps like https://plus.google.com/108243663090085262773/posts/gmHwiRxKNxJ, https://plus.google.com/108243663090085262773/posts/W5Nic8PcrFS [15:56] aquarius, ^^^ [15:56] sergiusens, can you elaborate ? [15:56] we ask him all the time :) [15:57] nik90: does he have tutorials? or just apps? [15:57] * mhall119 probably asks aquarius for too much already [15:57] mhall119, fine with me [15:57] thanks rickspencer3_ === rickspencer3_ is now known as rickspencer3 [15:58] mhall119: he just posted videos of his apps..maybe should request he write tutorials [15:59] * aquarius grins [15:59] nik90, that sounds like a good idea [15:59] nik90, but I'd watch out for that Langridge person. He's a bit weird [15:59] handsome guy, mind. [16:00] lol [16:00] aquarius: :) u playing with my mind..hehe...with nicknames [16:00] * aquarius grins [16:00] nik90, to be serious for a moment, dpm has asked me about six times to write some tutorials :) [16:00] I plan to do something when I get a chance, indeed [16:01] nik90, I told you we had :) [16:01] dpm: that cookbook page is looking much nicer now :) [16:02] mhall119, why, did you add a picture of aquarius to it? [16:02] dpm: hehe...adding more pressure always help :D [16:02] Uhm, this should be based on Qt, so it should work on Ubuntu Touch, no? https://twitter.com/mairin/status/322377455186890753 [16:02] Seems like something we should package and see :-) [16:02] dpm: not that much nicer :) [16:02] tedg: in theory [16:03] mhall119, Make it happen! :-) [16:03] or should I say, "in theory" with scare-quotes [16:03] are the screenshots deliberately blurred? :( [16:03] Yeah, I'm not sure why. [16:03] it depends on whether it's using components that we don't have [16:04] But apparently they're presenting at LGM as we speak. It might be a capture from the stream. [16:04] and it would obviously not follow the design guidelines, which is bad [16:04] aquarius: maybe they took a picture of their screen witha phone [16:04] then printed it out [16:04] scanned it back in [16:04] uploaded it to a photo sharing site [16:04] and spilled a glass of water on it [16:04] aquarius, Hah, like those will hold up past v1 ;-) [16:04] downloaded the thumbnail [16:04] :P [16:04] and sent via regular post [16:04] upscaled it [16:04] ... [16:04] but in theory it's doable, at least [16:04] mhall119, nice work with the cookbook page [16:04] in theory everything is doable [16:05] It's doable, we just have to figure out: how can we get enough beer and cigarettes to aquarius' house to make it happen? [16:05] * tedg checks amazon [16:06] amazon won't deliver cigarettes, I don' tthink :) [16:06] he doesn't smoke real ones anymore anyway [16:07] can you download more electronic cigarettes? [16:07] yeah. you could just have delivered a canister [16:07] full of that stuff that tastes like old socks if you burn it [16:08] or a big box of nicotine patches :) [16:29] seb128: cyphermox: https://code.launchpad.net/~tiheum/ubuntu-themes/new_icons/+merge/154470 is ready to be merged :) [16:29] kaleo, thanks [16:31] sergiusens: nvidia hack was at hybris [16:32] but still, the same version, weird [16:33] dpm: trying to create a spec and put it on ubuntu wiki but can't login, you seeing this at all? [16:33] bfiller, let me try. In the meantime, have you tried to log out from SSO and log back in? [16:34] dpm: let me check, don't think I was logged in to begin with [16:34] bfiller, I can still edit pages, but I was already logged in [16:35] ping mhall119 [16:36] cool cool, thanks kaleo [16:36] jenkins will merge it shortly [16:36] bobweaver: pong [16:37] dpm: I'm in, had to log out first. thanks [16:37] bfiller, cool, glad it worked [16:38] hey mhall who should I talk to about vm intergration I also made new templets for it [16:38] and some extrenal tools === dandrader|lunch is now known as dandrader [16:40] mhall119, maybe you would like to try the new template. ... I will make a video [16:41] bobweaver: I emails bzoltan your video, he's going to take a look [16:42] rsalveti, sergiusens, ogra_ Thank you guys, I am going to investigate more deeper [16:44] ogra_: and what exactly are these images? http://cdimage.ubuntu.com/kubuntu-active/daily-preinstalled/current/ [16:45] mhall119: I already did and pinged bobweaver [16:45] shadeslayer, broken crap i forgot to remove :) [16:45] hah [16:46] we tried to do n7 builds of kubuntu-active ... [16:46] yeah, I don't think kwin talks to SurfaceFlinger :P [16:47] no, that was desktop based [16:47] bzoltan, I am making a video of what I made should be done in ten minutes [16:47] ah [16:47] and when you say desktop based, you mean like the AC100 steps that you explained earlier right? [16:48] bobweaver: OK [16:48] shadeslayer, like the nexus7 desktop image [16:49] rsalveti, sergiusens Can you look at diifs between builds 20130303, 20130304, 20130305 and 20130306 ? [16:50] no idea what that is 0.o [16:51] http://cdimage.ubuntu.com/daily-preinstalled/current/ [16:52] ah okay [16:56] I have to go, can you mail me diffs to pb@P3T3.org ? Thanks [16:57] Saviq: https://code.launchpad.net/~fboucault/unity/phablet-use_icon_theme/+merge/158426 [16:57] Saviq: it's not quite ready but it's coming real soon [17:01] hmm, todays image the OSK doesn't work when i want to type a WPA key for network.. known issue? [17:02] on nexus 7 [17:03] keyboard doesn't appear [17:03] popey, did you try rebooting? [17:04] that was a bug that slipped in a week or so ago [17:04] every few reboots, you get that [17:04] rickspencer3: clean boot after new install, will reboot [17:04] popey, yeah, I bet it works after you reboot [17:04] ta [17:05] mhall119, hey, I notice that on the lenses, when you scroll up, the title smoothly scrolls off the top of the page [17:05] should I be setting up my apps in some way so that htey do that? [17:05] kaleo, ^ ? === SkavenXXI is now known as SkavenXXI-[OFF] [17:06] nope, reboot still shows no keyboard [17:09] rickspencer3: if I understand correctly what we are talking about, it should be automatic [17:09] rickspencer3: if you are using a MainView and some Flickable/ListView inside it [17:09] bah, reboot again it works [17:10] kaleo, I am, but I wonder if I have it set up wrong [17:10] it goes MainView{PageStack{ListView{}}} [17:10] rickspencer3: we have documentation now! :) [17:10] rickspencer3: http://developer.ubuntu.com/api/ubuntu-12.10/qml/mobile/qml-ubuntu-components0-mainview.html [17:11] rickspencer3: with examples [17:11] * rickspencer3 looks [17:11] rickspencer3: hmm, PageStack, let me see === jhodapp|lunch is now known as jhodapp [17:12] rickspencer3: doc is unclear about that case [17:12] kaleo, so the flickable needs to be a direct child of the MainView for this to work? [17:13] kaleo, oh, I lided [17:13] lied, even [17:13] it goes MainView{PageStackPage{{ListView{}}}} [17:13] dang it [17:13] rickspencer3: :) [17:13] it goes MainView{PageStack{Page{ListView{}}}} [17:13] rickspencer3: I would expect that to just work but the doc does not say it [17:13] rickspencer3: let's check with timp [17:13] I NEED MORE RAM !!! stupid thing keeps on frezzing [17:14] take 4 bobweaver pulls his hair out ep 3 [17:14] * ogra_ sends bobweaver a swapfile [17:15] also installing zram-config will help ;) [17:15] (will magically add ~1/3 extra ram to your machine) [17:15] rickspencer3: in the toolkit demos there is a similar case that works; let me look at the code [17:16] rickspencer3: yeah it is the same structure: MainView{PageStack{Page{Flickable{}}}} [17:17] rickspencer3: and it just works [17:17] weird [17:17] rickspencer3: can I test your code? [17:17] yeah [17:17] kaleo, let me see if I pushed it anywhere [17:19] lp:~rick-rickspencer3/+junk/technews/ [17:19] kaleo, ^ [17:19] thx [17:19] kaleo, aaah, I lied again [17:19] there is a rectangle between the listview and the page [17:19] I bet that is the problem [17:19] I need it but I hate it gstreamer !! [17:19] rickspencer3: :) [17:20] the rectangle is unnecessary, I think [17:20] kaleo, so the rule is, if the child of the page is a flickable, the autoscrollng will work? [17:20] rickspencer3: unless you want to display a colored rectangle then it is :) [17:20] rickspencer3: yes [17:20] take 4 [17:20] I don't even know why I put them in rectangles, tbh :/ [17:21] seemed like the thing to do at the time :) [17:21] because circles are trademarked by google ? [17:21] ogra_, yeah, probably something like that ;) [17:21] rickspencer3: fixed! [17:21] :) [17:22] thanks kaleo [17:22] I'll fix it up and blog about it later [17:22] ogra_: oh btw I can't find this on the interwebs, how does on start the adb daemon on the device? [17:22] in ubuntu touch it runs by default [17:22] nothing to start there [17:22] sure, but what exactly is that called? [17:23] ? [17:23] its adbd and android fires it up on boot [17:23] mm .. right, and adbd isn't something that's available in ubuntu [17:25] i'm not sure if the adb package contains the daemon ... never looked at that [17:25] but we usually dont use adbd in ubuntu, yeah [17:25] well, since I'm debootstrapping a basic install, once I put it on the device, how do I connect to it [17:25] which is why I needed adb [17:26] Allright got it that time Uploading it now bzoltan and mhall119 [17:26] or rather adbd [17:26] well, do you base on touch now ? [17:26] then your android layer has adbd # [17:26] and runs it on boot [17:27] hmm, yeah, but I distinctly remember my TF101 didn't have android bits [17:27] and it had adbd [17:27] no idea then [17:27] i know we have it in ubuntu touch by default [17:27] in the android layer [17:27] yeah [17:28] lilstevie: ^^ [17:28] lilstevie: I distinctly remember the ubuntu chroot you provided for the TF101 had adbd [17:28] bzoltan, and mhall119 video will be up in 5 minutes or so http://www.youtube.com/watch?v=JARd_eMYqOA [17:28] notice the new tools [17:29] like the Ubuntu Virtual package (adding in debian stuff today ) and the tools to launch the VM [17:29] Need to fix the vm and get rid of all the bloat though [17:30] add some upstart scripts ect [17:30] ah well [17:31] I'll just make it use wpa_supplicant and connect to my network [17:38] bzoltan, mhall119 video is done uploading let me know what you think [17:38] well besides that the vm needs work that is [17:39] for the wizards I see that you all are using json why not the reg way of xml? there are many more options and better ways to handle filenames [17:40] jppiiroi1en, ^^^ [17:41] as far as launching the VM I was thinking about putting that in with the cpp that is there for plugin already (ubuntu.pro) but I did not know if I should make my own button and screen do I just put it in extrenal tools for now [17:42] sergiusens, so looking at that image test page on jenkins, i see a ton of app tests, but how do i know from that if the image is good to go or not ? [17:43] at any rate I do not need to plug in my nexus 7 and make sure that it is on touch and what not I can now just use the virtual machine to test Hud and what not [17:44] if qsortproxyfilter would be more nice and application plugin was there then I could integrate this fully into desktop [17:44] sergiusens, do you have any toplevel thing i can poll (image good/image bad) instead of having to screen scrape the whole list of tests ? [17:45] seems like there needs to be more work done in that area but I have no clue who is in charge and what is and is not allowed so I am not touching that with ten foot pole [17:45] ogra_: I can probably right something [17:45] and right it in as a file [17:46] yeah, that would help [17:46] s/right/write/ [17:46] currently it seems a lot of the app tests still fail [17:47] ogra_: that's an autopilot bug [17:47] ogra_: need to fix today [17:48] well, if i would rely on the table at the bottom i wouldnt publish ... i can indeed easily just check if there are no fails at all but i would expect that we allow some level of fuzzyness [17:54] is aptdaemon running on ubuntu-touch platforms, or is there now python bits at all? and if it's not there, how does an app request that a package be installed/updated, or how does the user change mirrors or such? [17:58] so bzoltan and jppiiroi1en and who ever If you want me to do this just say yes or no. I just dont want to waste any time if it will never be used. As I have what I need atp but I could make real professional if need be But I would like to do a google hangout with the people that are in charge of this stuff so I DONT waste my time like I did With Ubuntu TV thanks [17:59] dobey: apt, apt-utils and apt-transport-http are on the phone, not much else in regards to application shopping [18:00] bobweaver: As I saw in the video you start a regular Ubuntu in a VM and launch the unity-next as a standalone app in it. I do not see the added value compare to just starting the QML apps on the desktop. [18:00] popey: is everything run as root? [18:00] dobey: by "everything" do you mean apps? if so, no, there's a non-root user 'phablet' [18:00] bzoltan, you can use Hud in normal desktop ? you can tie apps into touch on normal desktop ? [18:01] you can use on windows and mac on desktop ? [18:02] bobweaver: using HUD is a good one. [18:03] bzoltan, the idea would be to make the standalon app not stand alone and a shell. And make upstart for it make user "ubuntu" no login and kill unity and all that on start up [18:03] popey: how does "phablet" request that something be installed? you just have to open a terminal and run "sudo apt-get install foo" or something? is there a plan to have aptdaemon there, or are plans leaning toward something else? [18:03] bobweaver: Sorry, I missed the part of tieing apps into touch. Where was it? Why not to start the next without the VM? [18:03] bzoltan, that is the idea I only started this yesterday [18:03] dobey: we don't have that functionality yet [18:03] aiui [18:03] bobweaver: but it was not in your demo [18:04] dobey: I think aquarius is the best person to ask [18:04] i was afraid you might say that :) [18:04] :D [18:04] bzoltan, because that is not implanted as of yet. I would need to hack unity-next and if I am going to do that I need / want to make sure that what I am doing i the correct steps [18:05] no need to re-create the shell like I did with unity 2d if it is not what Canonical wants [18:05] so what I am saying or trying to lol is that I do not wnt to hack on the shell and the virtual machine iso if there is no reason to :) [18:06] bobweaver: What I am interested in is 1) a VM running the phablet image 2) qml-shell running in a nested display server [18:06] +1 to a VM running phablet [18:07] popey: yes, that is real deal. [18:07] my plain of attack is as follows 1) talk to you all. 2) hack the vm so that it i lean mean fighting machine. make the upload and paths and what not correct for the shell [18:07] 3) make qtcreator have more options [18:07] bobweaver: You talk to me, so that is a good start :) [18:07] 4 ) add button to launch the vm under the devices button [18:08] add tests to the wizards fo debian packaging [18:08] bobweaver: I am frank with you. I think you attack on too many fronts. I personally prefer simple steps and properly done small features. [18:08] Hi frank :) j/k I am glad that you are that is the best ! thanks [18:09] bzoltan, so the vm would need to be packaged with the sdk in the end with its own ssh keys and what not [18:09] bobweaver: I know [18:09] bzoltan, what do you think about that ? [18:10] bobweaver: The best would be native VM with the armhf phablet image in it. [18:10] still yet to see a reason why we can't use the android emulator [18:11] davik > [18:11] ? [18:11] like adk ndk style ? [18:11] bobweaver: the second option would be to start the qml-shell in a nested display manager [18:12] bzoltan, it is cgwinn chrooted ? for stock android ? [18:12] alan@deep-thought:~/android-sdk-linux/tools$ file emulator-arm [18:12] emulator-arm: ELF 32-bit LSB executable, Intel 80386, version 1 (SYSV), dynamically linked (uses shared libs), for GNU/Linux 2.6.8, not stripped [18:12] bobweaver: ^^ that one [18:12] bobweaver: yes [18:12] popey, that is fine but how to talk to qtcreator [18:12] adk ndk ? [18:13] via adk [18:13] adb etc [18:13] and what use ant for java [18:13] java? [18:13] bobweaver: qtc is a little problem [18:14] so is davik or how ever you spell it [18:14] adk ndk part ^^ [18:14] dalvik [18:14] Anyone encountered the QQmlComponent: Component is not ready bug? [18:15] becuase it is cgwinn or how ever uyou spell that :) [18:15] bobweaver: there are two options: 1) emulated platform with a full stack 2) emulated display server and shell [18:16] bzoltan, do you think that it would be better for vm stuff to wait for necessitias to get to qt5 ? [18:16] seems like bogden or what ever his name is is working on that [18:17] bobweaver: what is missing from qt5? [18:17] compiler and a couple of other things [18:17] phone brb [18:18] bobweaver: compiler? I do not understand... [18:18] Rttommy: yes [18:18] for a7 [18:19] also I do not think that minestro has any repo for qt5 [18:19] why woudl you need a compiler in a vm ? [18:19] for apps [18:19] you cross build already [18:19] netcurli: How can I fix it (I installed the SDK as instructed, on 13.04) [18:19] no need for it to live in the vm [18:19] ogra_, Oo [18:20] you only want the vm for executing your binaries and test them [18:20] no for qt I am saying [18:20] (at least we do) [18:20] ogra_, have you ever used necessitia ? [18:20] compiling inside the vm will be slow as hell [18:20] no no no [18:21] it compiles and sends package to device [18:21] no, i havent even ever used Qt ... but i have worked with arm vms for years [18:21] adt or whatever they are called [18:21] right, and we would like to replace the device with a vm [18:21] android app like deb but for android [18:21] but keep the setup beyond that as is [18:22] Hi I'm trying to build cm10.1 for padfone 2 anyone interested in helping me please. [18:22] currently we do cross builds that we push to the real device running UTouch [18:22] from the SDk [18:23] nt? [18:23] and we would like to have the VM for people that dont have a device [18:23] no other changes [18:23] so we need a VM that runs UTouch [18:24] and that the SDK can push to [18:24] Rttommy: you are only talking about the error message, right? the application runs fine otherwise? [18:24] so necessitia was made so that qt developers could make apps on vm to test (or to deploy to device ) if no device it launchs emulator (dalivik ) and then makes package for android it then after adb pushing used necesstia to call minestro that looks at device if package is(qtcore / whatever) is installed your app launchs (this happens only once ) if it is not it installes them [18:25] project lighthouse I thinkit was called [18:25] netcurli: no, many elements (in the tutorial and the touch core apps) do not render correctly. [18:25] Think is key word [18:25] bobweaver, right we would want the dalvik part without dalvik but instead with UTouch :) [18:25] correct [18:25] that is the hard part [18:26] Rttommy: do you get other error messages as well? [18:26] one idea was to have an x86 android build that runs in qemu [18:26] same thing we run on the devices but x86 and qemu based [18:26] netcurli: the only output is : QQmlComponent: Component is not ready /usr/lib/x86_64-linux-gnu/qt5/bin/qmlscene exited with code 0 [18:26] that is what I have in virtual box ogra_ but I just did not want to take all the time to fix it if not a good idea [18:27] bobweaver, so you did an android x86 build ? [18:27] it is concept [18:27] ogra_, yeah [18:27] of our device tree ? [18:27] things are missing [18:27] device tree ? [18:27] pahblet.ubuntu.com [18:27] *phablet [18:28] our stripped android base [18:28] I compiled it and had to alter some tests and what not Hud was a b&*^( [18:28] Hud ? [18:28] i'm talking about the device layer ... [18:28] not the rootfs [18:29] no this is just concept:( but the idea if x86 then take the ubuntu-minamal and start there build a working VM [18:29] bobweaver: I do not think we want a VM based cross compiler in the Ubuntu Touch SDK in the following half a year. [18:29] Rttommy: and what exactly is not working? can you pastebin the code you try to run? [18:29] we want to emulate the device layer ... [18:29] a great example of this would sailfish and how they are doing it [18:29] bobweaver: for armhf target I suggest to use the real device or crosscompile on the desktop [18:29] but with out mer [18:30] netcurli: https://bazaar.launchpad.net/~dpm/+junk/CurrencyConverterUpdated/files/8 [18:30] bobweaver, the issue is that we need the platform-api and libhybris bits running in android to have proper emulation [18:30] bobweaver, else it is just Ubuntu Touch being run in an x86 VM [18:30] we want to emulate the whole stack [18:31] bzoltan, I agree seems like you all have the device stuff under wraps all ready for the creator [18:31] including the lowest layer [18:31] ogra_, How to tie into qtcreator ? just wondering ? [18:31] so Mir can eventually just blend in (and with luck directly talk to the deslktop Mir) [18:31] Rttommy: and it doesn't look like the screenshots from the tutorial? [18:32] bobweaver: what we want is what popey, ogra_ and me are saying... either emulate the whole stack or just run the shell in a nested display server on x86 [18:32] right [18:32] what bzoltan said [18:32] like xephyr ... but using Mir :) [18:32] how far has the terminal app yet? [18:32] ogra_: exactly [18:32] or a full android VM [18:32] yeah I am onboard for that bzoltan Just wondering If I should start on the second one ? [18:32] *is [18:33] bobweaver, thats hard since there is no Mir for the desktop yet [18:33] zephyer is AWESOME ! [18:33] bobweaver: That would be a great idea [18:33] we would need one mir to talk to the other [18:33] correct [18:33] bobweaver: yes, I liked zephyer from the N900 and N9 SDK :) [18:33] and to date only the android Mir is working afaik [18:33] netcurli: Nope, the header is empty, and part of the main area is hidden under the header. [18:34] ogra_, / bzoltan x11 VM super striped down that only has shell running ? [18:34] nope [18:34] that will not work :) [18:35] ogra_, or bzoltan google hangout ? [18:35] Rttommy: then maybe that has to do with the tutorial code not having been updated for the current UI toolkit version [18:35] nope, i had a full day of hangouts today ... (its late here) ... we can do one tomorrow if you like [18:36] bobweaver: Only if you teleport to Finland in the unfortunate case my kids wake up from my voice.... [18:36] bobweaver, the point is that what you know as shell today will soon make full use of Mir features [18:36] I would like to start working on x86 striped down one [18:36] cool [18:36] so X11 wont help [18:36] netcurli: That's what I'm starting to guess, thank you, I'll wait until the code is updated [18:36] yeah I meant x86 [18:36] sorry [18:36] ah [18:36] mir is a no go on x86 ? [18:37] nope, but further from being ready than on arm [18:37] they focus on the android drivers only atm [18:37] so when S opens you will see Mir land for UTouch but not for the desktop yet [18:38] lilstevie: poke poke [18:38] what we need is something like http://www.android-x86.org/ ported to phablet.ubuntu.com [18:38] as our base [18:38] there Mir will just easily work [18:39] why not make it happen like that striped down x86 so the only thing that is running is the ubuntu-ui-toolkit example and also the shell (not as stand alon app though ) make dirs so thtat they are readable and have them install (qtcreator-->dirs) and have it restart unity on start up ? [18:39] the alternative is to wait until the desktop Mir lands ... but that will be to late [18:39] so that way app can talk and act in shell (phablt) [18:39] the prob is that we will start to rely on Mir functions [18:40] you need a way to run that bit [18:40] so scrap it and add mir to it ? [18:40] to what ? [18:40] from x11 [18:40] the vm ^^ [18:40] if Mir only riuns on androids libGLES atm [18:40] *runs [18:41] but later ? I mean it is going to run on x86 I would hope [18:41] the problem is that there is no way to replace X11 yet [18:41] yeah [18:41] later it will [18:41] which apps are working right now [18:41] but that will be to late for us :) [18:41] so x11 now then replace with mir [18:41] I mean the Virtual machines can be updated with package [18:41] as soon as S opens and the new unity lands it will start depending on Mir [18:42] I have virtual machine of zephyer for Mer [18:42] and will likely not be fully functional in an X11 based VM [18:42] I see [18:44] that is what I make all my qt5 apps for wayland on is the zephyer using Mer and qtcreator [18:44] would it not be the same with Mir ? [18:44] sorry just trying to get this all together ? [18:44] not sure [18:45] well, the think is that we will make heavy usage of the compositing of Mir in the UI [18:45] rsalveti, Do we have support for the light sensors on the current Nexus devices? [18:45] *thing [18:45] you wont be able to emulate that in X or zephyr [18:45] tedg: not yet, ChickenCutlass looked at that during the london sprint, but we didn't yet finalize the implementation [18:46] its trivial [18:46] (teh light sensors) [18:46] you can read their value from sysfs [18:46] Hmm, we were just curious about that for indicator-power. Whether it'd make more sense to use that data instead of a slider. === dandrader is now known as dandrader|afk [18:47] ogra_, here is my zephyer set up https://lh4.googleusercontent.com/-_y2zcbpF7cw/UTo9rKCGEXI/AAAAAAAAA10/D3DF5ANvmWo/s902/Screenshot+from+2013-03-08+14%3A34%3A38.png I just use that as testing bed for apps (alot of things dont work yet on it though ) [18:47] tedg, http://paste.ubuntu.com/5699366/ ... thats what i use on the chromebook ... i dont have my nexus7 handy buts its similar [18:47] ogra_, Is there a upower interface for that? I thought there was. [18:47] https://wiki.merproject.org/wiki/Zephyr [18:47] ogra_: getting the values is trivial, but we don't have an api for that [18:47] tedg: brighteness is different [18:47] you need to use sysfs indeed [18:47] rsalveti, right [18:47] like what is done by g-s-d [18:48] so I guess that there will be a qt-mir like qt-wayland [18:48] g-s-d doesnt do light sensors [18:48] but for the auto-brighteness you'd need that indeed [18:48] bobweaver, yeah ! [18:48] and that's why it's not yet supported even on phablet [18:49] but getting an api together shouldnt be hard ... work of a day or two [18:49] since you can already easily get and set the values [18:50] just needs the high level definition [18:50] and i think janimo started some work on that for the g-s-d nexus7 stuff for the desktop image [18:50] I'm surprised that Upower doesn't abstract it. [18:50] not sure if there is proof of concept code [18:50] Is the sysfs interface constant for everyone? [18:51] brightness is, lux/illuminance isnt afaik [18:51] tedg: at least for brightness it is [18:51] ogra_: doing the api is easy, the hard part is doing it in a way it's common and standard enough for us to use at somewhere else ;-) [18:52] we don't want to keep changing api all the time [18:52] yep [18:52] +1 [18:52] so that's why we decided no to do it until we have a standard way of handling all the sensors [18:52] Hmm, doesn't seem my laptop (which has a light sensor) has anything "illuminance" in there. [18:52] look for lux [18:54] i think the non standardization of the kernel interfaces is the reason you dont have that in upower and g-s-d [18:54] Uhg [18:54] its usually a platform device on arm and everyone picks their own names [18:54] not sure how it is on x86 [18:54] but i wouldnt expect much better there [18:55] even worse at the android based kernels [18:55] the illuminance interface should be identical on the nexus 10 btw [18:56] since chromebook and n10 have the same board afaik [18:56] that's why android has an abstraction layer for that [18:56] yep [18:56] * tedg still hasn't found it on his machine [18:56] heh [18:56] I think the only thing related with light sensor at upower is related with the keyboard [18:56] x86 HW is so weird and ilogical ... [18:56] :P [18:57] bzoltan, soorry to nag well.... nr but should I make proff of Concept x86 I mean it is going to be super bare bones [18:57] hey ogra_ touch version has no sound on a n7 is this something that should get fixed with the move to raring? [18:57] bobweaver: no worries... you have heard ogra_ about the possibilities [18:58] davmor2, known bug, not fixed in raring [18:59] ogra_: :( but know :) [18:59] davmor2, the android layer is the same in raring and quantal ... its a low level thing [18:59] ogra_: ah okay that makes more sense then [18:59] davmor2: ogra_ recomments audiobooks with subtitles to work around it :) [18:59] :) [19:00] ogra_: is it the same thing with the camera though I haven't tried that lately [19:00] bzoltan, ogra_ so just dont pull out qx11extras what ever I do then :P [19:00] mhall119: are they not just called books at that point ;) [19:01] davmor2, yeah, just draw signs with subtitles :) [19:01] davmor2: no no, way better [19:01] :) [19:01] haha [19:02] mhall119: I think you have worked with Europeans too long you seem to get sarcasm :P [19:02] mhall119: that jono has a lot to answer for :) [19:02] davmor2: I'm from the south, I was born into sarcasm [19:03] mhall119: isn't that in Louisiana [19:03] the dangly-part of the south [19:04] the norway of the equator ? [19:05] ha, I like that, I'm using it from now on [19:05] :) === dandrader|afk is now known as dandrader [19:30] hmm, qml shell seems to have barfed on my daily today [19:31] rsalveti, Okay, so it seems there's iio in the kernel that is supposed to be "the" way to do ambient light sensors. [19:31] in fact there are no processes running as phablet [19:31] rsalveti, There are a few drivers for it. Do you know what chips are in the nexus devices? [19:31] tedg: no, but can check [19:31] tedg: did you find which kernel subsystem this is in? [19:32] rsalveti, iio, which doesn't make sense to me. But, eh. [19:32] restarting qml-phone-shell now It cant see any networks [19:32] seems properly borked [19:32] * popey reboots [19:32] rsalveti, Apparently it's relatively new. [19:32] tedg: right [19:32] let me investigate as well [19:33] rsalveti, meet sforshee, he has taught me everything I know about iio :-) [19:34] :-) [19:34] geez, you must not know much then [19:34] cause I sure dont ;-) [19:34] That is a very good assessment :-) [19:35] rsalveti, sforshee was also interested in figuring out what we're doing for various other HW abstractions. [19:35] I don't know much about that either, but thought getting you two to talk would be best. [19:36] actually rsalveti and myself and some others are supposed to chat about it next week [19:36] as not everything at android is following the usual standards, and also providing stuff via binaries, we initially thought about abstracting the sensors interface via an api similar to the platform-api [19:36] so I believe we'd need some sort of api/abstraction layer [19:36] and hopefully get better drivers as well following whatever is standard at the kernel [19:37] yes, I'm of a similar opinion after looking at the drivers for a couple of devices === chilicuil is now known as chilicuil_away [19:37] for some things the kernel interfaces are well-enough defined to serve as the abstraction, like input for example [19:37] but for other areas like sensors, not so much [19:38] right [19:38] I mean, if we're going to invest in work there, would it make sense to do that work in the kernel to get better interfaces? [19:38] sure, but we could do that in parallel, as we cannot easily change the current drivers available [19:38] except we're targeting android devices which are already using ad-hoc interfaces into the kernel [19:39] remember we need a way to reuse what is already available as well [19:39] besides looking forward for the right implementation [19:39] Sure, but if I remember right there's a way to do some sort of loopback style thing in a kernel driver. [19:39] So we could mimic the interface. [19:39] for some types of devices [19:39] Sensors? :-) [19:39] we might have to invent it for e.g. iio [19:39] I doubt it [19:40] input has that [19:40] also we have to keep in mind that some devices don't have a kernel driver at all [19:40] gps on the nexus 4, for example [19:40] we also need to better understand how android is currently handling that, and how that's used at least for the devices we support [19:40] exactly [19:40] Hmm, that is much trickier. [19:40] they might only export the memory interfaces sometimes [19:40] or not even that [19:40] or be completely user space [19:41] i2c? [19:41] not positive, my guess is a uart [19:41] plus some gpios [19:41] I guess I'd just like it to be where we could use something like udev or upower to access them. [19:41] Not invent another interface there. [19:41] Just completed initial design for "Cellular" phone settings. https://wiki.ubuntu.com/Networking#Phone [19:42] tedg: I'm not sure that this will be possible [19:42] we might end up needing another abstraction layer [19:42] rsalveti, :-( [19:42] I was hoping we could just tell udev that it's an iio device if that's what we think the final would be. [19:43] that's why the idea was to follow and extend kind of what is currently done at the platform-api [19:43] Even if udev is doing something crazy in userspace. [19:43] we cannot enforce this final decision either [19:43] we can enforce as a distro, but not when doing a product [19:43] tedg, what might be realistic is that we provide abstraction libraries for the "standard" interfaces to make something like iio work automagically [19:44] and someone making a product could supply their own libraries [19:44] we'll kind of get whatever the vendor is able to give us, and depending on the timeframe, we might not have time to do it the 'right' way [19:44] exactly [19:44] that's exactly the big and most useful thing from android [19:44] the abstraction layers, even if that makes a pita to maintain from a distro perspective [19:45] but working with oems is not an easy task [19:45] and android succeeded on that [19:45] Sure, I guess I just feel like we have an abstraction layer already. Let's put more in there than build another one. [19:45] in an ideal world ... [19:46] well, we're redesigning the current one, which is the platform-api [19:46] Let the vendor library link in there, not into the user space. [19:46] right [19:46] the platform-api is abstracting the access via hybris (and then android libs) [19:47] Well, platform API is kinda silly. I mean, it's hard to use because it's all pure C/C++. Where things like udev have simple libraries already built to access them. [19:47] right, that's why I said *we* can propose something better now [19:47] while we kind of redesign platform-api [19:48] and I believe that was the work ChickenCutlass was kind of responsible for [19:48] rsalveti: what did I do now ? [19:48] lol [19:48] I guess my thought would be that window and application management need to be separate from sensors/drivers. [19:48] That stuff has to be in user session, where the other is system level. === highvolt1ge is now known as highvoltage [19:49] ChickenCutlass: discussing about the abstraction around the sensors :-) [19:49] right [19:49] tedg: I'm +1 on that [19:49] I'd like to abstract that at a base os/platform layer [19:49] +1 from me too [19:50] separated from ui and session handler [19:51] Trying to think of why it bothers me. I think it's because one needs to go across user boundaries, and one doesn't. [19:51] There'll always be one rotation sensor, but there could be a session manager per suer. [19:52] (well, there could be two axis, but I think you get the point) [19:52] ogra_: okay, last step remaining, I have no idea how one creates an initrd [19:52] and sometimes we'd also need to abstract to reduce sensor usage, in a way only the abstraction/provider would control the real access to the sensor [19:53] like, we don't want to have a bunch of apps taking the data in parallel directly from the hardware [19:53] but instead trying to sync that at a provider [19:53] rsalveti, Are you thinking just for accessing it at the same time issues, or power? [19:53] I think both are valid concerns [19:53] yeah [19:54] if all drivers are completely in the kernel then we _might_ be able to deal with concurrent access [19:54] rsalveti: so currently there is a sensor service on the android side that controls access to the data. [19:54] but I've seen drivers which do power control by exporting a gpio to userspace [19:55] so if you have multiple processes trying to control power, things are likely to go bad [19:55] right, it's all kind of crazy, lack of standards all around [19:55] ChickenCutlass: right, we might end up using that or doing something similar to it [19:55] rsalveti: yup [19:56] I think the key is that we need something that works for laptops too. If we use it, we need to figure out how to make it work on that side of things. [19:56] Not sure if that's a big deal or not. But I really want my backlight to use my ambient light sensor ;-) [19:57] sure, we need to abstract all use cases [19:58] So, what triggered all this is mpt asking whether we need a slider for brightness in the power indicator or we could solve this with ambient light sensors. [19:58] And it seems like there is generally support for it, but we're not quite close to making use of it. [19:59] well, we need both [19:59] we need auto-brightness and to allow the user to manually set the level of brightness [20:00] rsalveti: if you apt-get update the raring image, it will be mostly functional [20:00] sergiusens: cool, let me try [20:00] rsalveti, Because of ambient light sensor not being accurate enough, or because we don't have support for it? [20:01] For instance, iOS doesn't expose it easily. [20:02] tedg: well, android exposes both, so I just assumed that [20:02] I'd hate if the device didn't allow me to control brightness [20:02] just because I personally hate when the device tries to be smarter than me :-) [20:02] Heh, it's like the number of people at UDS that complain we don't have a good CPU monitor. [20:02] :-) [20:03] tedg: but I agree, that's easier to assume at i* devices [20:03] because they fully control the hardware [20:03] so they can make sure the sensor is good enough at most of the cases [20:03] which is not the case for android based devices, and which would be similar with us I believe [20:04] and can we really assume that all devices even have an light sensor? [20:04] There is probably a chicken and egg problem there, eh? The OEMs don't work on it because they figure users adjust it. [20:04] probably not [20:06] I would think so for mobile... [20:06] I mean, laptops, no. But every phone would need at least a proximity and a light sensor. [20:07] yup, but if we allow the user to manually control the brightness level that would work for all cases already [20:07] so that's why I think we should have both [20:07] It seems to me like something that could be in settings instead of the indicators though. [20:08] tedg: the entire idea of the touch indicators was to allow the user to change stuff there directly ;-) [20:10] Heh, I wish we could have the few of settings. [20:10] that few [20:10] Would make my job easier! [20:11] :-) [20:23] could anyone explain to me how the initrd.img for the boot.img is made? [20:23] and if possible, which kernel is used ? [20:24] shadeslayer: they are built as a build step from the android image currently [20:25] using the kernel sources from android as well === seb128_ is now known as seb128 [20:25] depending on which device you built against [20:25] okay, and what if I want to build a initramfs that can boot ubuntu but use the android kernel, is that possible? [20:25] shadeslayer: yup [20:26] could you give me a quick run down? :) [20:26] ChickenCutlass: might have a few pointers easily, he is mostly doing the same thing at this moment [20:27] aha [20:27] :) [20:27] abootimg helps you extracting and creating the boot.img [20:27] yeah I have that part down :P [20:27] so you can extract that, put your own initrd and create it again [20:27] but I have no experience in creating an initrd [20:28] I've always downloaded premade initrd's and they worked :P [20:28] right, that's done by the update-initramfs tool available at ubuntu [20:28] anyone here workign on a one x port ?? [20:29] rsalveti: what I'm trying to accomplish : Android kernel + any initrd that can boot ubuntu + a debootstrapped rootfs in userdata [20:29] rsalveti: then I can pass rootfs=/dev/block/whatever and I think it'll work [20:29] as i ntoiced the current maintainers seem to be mia [20:29] atleast that's how the TF101 worked [20:29] sergiusens: awesome, lens working again :-) [20:30] sergiusens: how far are we from fixing the resolv.conf issue? [20:30] shadeslayer: right [20:30] rsalveti: yup... I'll double check with tim... if not, I'll workaround it [20:30] sergiusens: ok, so we can have a fully working build tomorrow already [20:30] oh, that sounds right? [20:31] shadeslayer: usually there's not much hardware specifics at the initrd [20:31] rsalveti: yeah [20:31] rsalveti: I wrote my own [20:31] you could have modules added by ubuntu, but you can remove that [20:31] i dont no what device oyur workign on btu whats teh problem with resolv.conf [20:31] but it dodn't work :P [20:31] it keeps rebooting [20:31] shadeslayer: :-) [20:31] crashing somewhere [20:31] humm resolv.comf is cuasign a system reboot or dchcpcs [20:31] cd* [20:32] rsalveti: followed this http://www.ibm.com/developerworks/library/l-initrd/ [20:32] ChickenCutlass: do you have the initrd you used as reference from the desktop image for nexus7? [20:32] we can just use that [20:32] though that guide doesn't put a init call anywhere [20:32] right [20:32] otherwise would need to grab from my pand [20:32] pnda [20:32] ^_^ [20:33] why do I need ubuntu for android rather than runnning android with some chroot goodness? [20:33] ulph: im assumign becuas ubuntu is runnign from an android base btu is usign its own packages and i think its sitll rmeotely debian based [20:33] remotely* [20:34] I have everything ready, I just really need a initrd [20:34] :P [20:34] shadeslayer: erm what you need initrd for [20:35] nubby: the idea is to use a ubuntu initrd + Android kernel and a debootstrapped ubuntu in userdata [20:35] then I can get rid of surfaceflinger and use X + KDE [20:35] ahh what device [20:35] not to mention other android stuff is also removed [20:35] Nexus 10 [20:35] ahh [20:36] wishign the dev team would do some work on the one x port [20:36] im to lazy to get the itnial boot working [20:36] I have an HOX [20:36] but heh im pretty sure the lakc oif networkign is ovrsite on ther part [20:36] but I don't want to screw my only phone ;) [20:36] heh you shoudlent scrfew it the oen x is prety mcuh unbrickable [20:37] as logn ads oyu have recovery and a kernel installed you should be safe [20:37] well, I meant, it's my only phone, and if I isntall unstable things, I won't be able to make calls or whatever [20:37] fair point [20:37] shadeslayer: right, let me boot my panda [20:38] awesome [20:38] shades you lucky enogh to have an s-off'ed [20:38] nubby: no idea, its all stock [20:38] ahh [20:38] lols [20:38] s-off'ed hoxs are rear as rockign horse shit [20:39] unless you got a the lte version [20:39] then there ment to be an s-off method sicne it isnt based on tegra 2 [20:39] tegra3* [20:40] rsalveti: I don't suppose you have an idea on how big the Nexus 10 boot partition is :P [20:40] shadeslayer: can try to check [20:41] okay [20:42] rsalveti: fwiw I made a ramdisk using mkinitramfs -o ramdisk.img in a armhf chroot [20:42] bootloop [20:42] how big is the source for ubuntu-touch and all its nessercery bits and peices [20:42] :( [20:44] shadeslayer: http://people.canonical.com/~rsalveti/initrd-raring.img [20:44] * shadeslayer tries [20:44] gzip -dc ../initrd-raring.img | cpio -ivd [20:44] find . | cpio -o -H newc > ../initrd-new.img [20:45] + abootimg should be enough [20:45] erm why extract it? [20:45] in case you want to change it [20:45] ah [20:45] lets just try it [20:47] mmm nope [20:48] uhh [20:48] rsalveti: can you md5sum that for me? [20:49] d68230a5edb8ca965ca3d46c7cffde25 initrd-raring.img [20:50] okay [20:51] crap, my n10 is out of battery [20:51] herp derp [20:51] not in gzip format [20:52] rsalveti: shadeslayer I just installed a kernel in my armhf chroot and grabeed that initrd [20:52] ChickenCutlass: and it worked? [20:53] shadeslayer: not yet [20:53] heh :P [20:53] shadeslayer: working onit [20:53] awesome [20:55] ChickenCutlass: any ideas on how big the boot partition of the N10 is? [20:56] shadeslayer: I don't know off hand [20:56] I've previously faced issues with the TF101 where the bootpartition was too small [20:56] and the boot.img I flashed was too large ( the flash command returned successfully ) [20:56] shadeslayer: should be big enough for a minimal initrd [20:56] well [20:56] I /think/ it's 4 MB's [20:56] from the android boot img [21:01] well now I know [21:01] it's 16 MB's [21:01] plenty big [21:01] http://paste.kde.org/722570/ [21:01] yep ^^ [21:02] the TF101 had 5-6 MB's I think [21:03] http://pastebin.ubuntu.com/5699761/ [21:04] lrwxrwxrwx 1 root root 20 Apr 11 20:28 boot -> /dev/block/mmcblk0p6 [21:04] BOARD_FLASH_BLOCK_SIZE := 4096 [21:04] you have 22 MB's? 0.o === chilicuil_away is now known as chilicuil === rickspencer3_ is now known as rickspencer3 [21:06] rsalveti: http://pastebin.ubuntu.com/5699768/ [21:06] shadeslayer: yup [21:06] quite different ... hmm === salem_ is now known as _salem [21:06] shadeslayer: which device? [21:06] also, I'm using 4.2.2 [21:06] N10 [21:06] ah [21:06] 4.0.4 should be different [21:06] well I think I updated it earlier today [21:06] when I got it [21:06] so should be 4.2.2 [21:07] but then I flashed ubuntu on it [21:07] I don't even have dw_mmc.0 here [21:07] shadeslayer: what is the emmc size of yours? [21:07] mine is 16gb [21:08] that probably made different partitions I guess [21:08] same [21:08] and the controller [21:08] oh crap [21:08] controller? [21:08] I'm doing that at nexus 4 [21:08] lol [21:08] too many devices plugged in [21:08] clearly ;) [21:08] shadeslayer: now it matches yours [21:09] :P [21:09] ^_^ [21:13] http://www.reddit.com/r/XXXBITCHINBEACHXXX/comments/1c5t0e/beautiful_girl_big_boobs_beautiful_face_wonderful/ [21:15] what the fuck [21:15] Please mind your language [21:15] sorry :S [21:15] All #ubuntu-* channels follow the same !guidelines please read them if you are not aware of them [21:16] !guidelines [21:16] The guidelines for using the Ubuntu channels can be found here: http://wiki.ubuntu.com/IRC/Guidelines [21:16] I see nothing [21:16] http://www.reddit.com/r/XXXBITCHINBEACHXXX/comments/1c532e/worlds_stronges_vagina_guinness_book_of_records/ [21:16] IdleOne: ^^ [21:16] .... [21:16] :/ [21:16] sorry I didn't see that before [21:20] IdleOne: doesn't make me right ;) [21:20] true :) [21:23] !ops [21:23] Help! Channel emergency! (ONLY use this trigger in emergencies) - elky, Madpilot, tritium, Nalioth, tonyyarusso, PriceChild, Amaranth, jrib, Myrtti, mneptok, Pici, jpds, gnomefreak, bazhang, Flannel, ikonia, maco, h00k, IdleOne, bkerensa, nhandler, Jordan_U, DJones or k1l! [21:23] yeah, that works [21:23] needs to be edited [21:23] ignore that [21:24] * k1l looks in [21:24] sorry [21:26] !no ops-#ubuntu-touch is Help! Channel emergency! (ONLY use this trigger in emergencies) - oubiwann, corp186, popey, k1l, Corey, wgrant, idleone, and ogra. [21:26] sweet! [21:27] !ops [21:27] Help! Channel emergency! (ONLY use this trigger in emergencies) - oubiwann, corp186, popey, k1l, Corey, wgrant, idleone, and ogra. [21:27] done [21:31] Not sure that oubiwann would want to be pinged there. [21:34] thanks IdleOne [21:34] sure thing [21:35] Anybody who doesn't want to be in that bot trigger feel free to PM me and I'll fix it [21:35] if I forgot anybody let me know also [21:40] ChickenCutlass: how does one even debug issues with the initrd? all I get is the google logo [21:40] and then it reboots [21:44] anyone ever see "protocol failure" when phablet-flash tries to push the zipfiles to the device? [21:44] I retried the adb push by hand, and it worked fine [21:47] shadeslayer: without a serial console it is flying blind. [21:47] heh [21:48] I don't suppose the N10 has a serial console muxed over USB [21:48] plars: yeah, but I only see it on raring [21:48] shadeslayer: you could possibly build the kernel with the USB gadget driver and try to get a USB serial console going. [21:49] ChickenCutlass: reading up on what a usb gadget driver is .. [21:50] that will be a major pita [21:50] oh [21:50] might be easier to patch in early framebuffer [21:50] sergiusens: this is on quantal desktop and latest quantal-preinstalled image [21:51] sergiusens: I've seen odd timing issues with adb not being *quite* ready after a wait-for-device before, but this is a new one, trying to reproduce now but I suspect it's a fluke [21:52] shadeslayer: http://www.samsung.com/global/business/semiconductor/file/product/Exynos_5_Dual_User_Manaul_Public_REV100-0.pdf [21:52] looks like it has the pin muxing details [21:52] protocol failure [21:52] Error while executing adb -s 0149BD7E0200E00F push /home/ubuntu/Downloads/phablet-flash/69/quantal-preinstalled-phablet-armhf.zip /sdcard/ [21:52] plars: might want to check you dmesg for usb resets or something similar... at least on raring, after that I get the annoying MTP message [21:53] ali1234: better share that with ChickenCutlass [21:53] I just got it twice in a row [21:53] sergiusens: nothing interesting in dmesg, but I can still run it by hand with no problems [21:53] he probably knows how to get this working better than me [21:54] shadeslayer: I will get it working. [21:54] shadeslayer: give me some time. [21:54] plars: hmmm... I'll check the script for weirdness, but I think I already have :-P [21:55] ChickenCutlass: right :) [21:55] I have no idea how to :P [22:10] looks like the USB pins aren't muxxed with anything useful... i don't think you can even GPIO with them [22:11] that sucks [22:12] okay, I'm sleeping, will pursue this tomorrwo [22:12] night :) [22:27] rsalveti: still around? [22:27] sergiusens: yup [22:27] rsalveti: have an MR for you [22:27] sergiusens: sure :-) [23:00] Is there an official Google blog for ubuntu touch somewhere? I want to know when the next release happens. === rickspencer3_ is now known as rickspencer3 [23:26] rsalveti, good raring tomorrow eh? what timezone are you in ? :p [23:30] Linton, there is a daily I believe. [23:31] https://plus.google.com/u/0/114111362741083777649/posts/1j5HvB1uqBT [23:31] tease [23:31] lol [23:31] I was missing the terms "daily" or "changelog". That's what I needed. Thanks!! [23:42] Help everyone [23:42] i MUST KNOW [23:42] what are the new lib hybris libs [23:43] my retarded little brother broke my android compiling laptop [23:43] i was wondering [23:43] can i just copy the new libs [23:43] dank101: https://github.com/stskeeps/libhybris [23:44] Compiled? [23:44] oh, sorry, I misunderstood what you were asking [23:44] Lol [23:45] Yeah pls