[00:13] Hi [00:13] Can Ubuntu Touch be compiled to run on an arm v6 processor? [00:16] I feel Ubuntu Touch could be a great success if it could be used to breathe life into older phones [00:16] I have an entry-level Android device from late 2011 and this thing is barely capable once a few apps are installed. [00:16] Is Ubuntu Touch noticably faster than Android 2.3.7? [00:22] alfonsojon: I've only run it on Android 4.x capable devices, so I'm not sure [00:23] Ubuntu won't work on armv6 though, it needs v7 [00:24] Darn. [00:24] I should save up for a Nexus 4. [00:24] :) [00:25] If I had money, I would contribute to the Ubuntu Touch in a heatbeat [00:25] but, I'm only 16 and failed at applying for a job. so meh. [00:32] sergiusens, i dont think it's a issue of space added a copy of the touch script on the boot.img and it worked succesfully if i left scripts/touch intact. [00:41] well ill talk to ogra since apparently if i change one line of scripts/touch it breaks [00:42] annerajb: how are you changing it? [00:42] that doesn't make sense [00:43] sergiusens, i tried everything used a sed on the target that compress the boot.img to change it and repack it. now i am copying the complete file from another place and repacking. [00:43] sergiusens, this is how it looks like now http://paste.ubuntu.com/5983196/ [00:44] annerajb: so you unpack, modify, repack? [00:44] yes [00:44] annerajb: can you provide a diff from the original? [00:44] yes [00:46] sergiusens, on the original i added only line 28 and it worked 70% of the boot's yesterday [00:46] this one has a bunch of kmsg gime a min for the diff [00:47] hmmmm [00:48] sergiusens, http://paste.ubuntu.com/5983206/ === Ursinha is now known as Ursinha-afk [00:50] sergiusens, i got it i am 99999% when i repack it it's missing +x permission [00:51] not sure how to set the user group for the initrd but ima set it to 777 and see if it works now === jhodapp|bbl is now known as jhodapp [00:57] annerajb: line 29 you mean? [00:57] annerajb: can't you pass in path to init directly instead of modifying? [00:57] sergiusens, path="/dev/mtdblock3" that I had only that yesterday and it worked all morning and after a while at night it stopped working [00:57] sergiusens, like a argument? [00:58] yeah, something like this https://github.com/janimo/android_device_huawei_u9200/commit/b3f839844fcb996ded1259a31d270236d3081c9c [00:58] but with path= [01:00] sergiusens, i still have to add another change talking to ogra he set that i need to add a sleep before my moutn because apparently udevdm settle is not working no my device [01:00] that would put me on the same state as yesterday where i have it booting but the moutn fails because it moutns too fast before device is ready [01:07] sergiusens, and to finish the night here is a booting dmesg with my touch script after formatting /data http://paste.ubuntu.com/5983266/ [01:08] [ 7.255078] VFS: could not find a valid V7 on mtdblock3. === n7_ is now known as Guest518 [01:17] i can finally install ubuntu touch on N7 [01:17] but i don't know [01:18] what is password root ? [01:19] Guest518: "phablet" [01:19] is the sudo password for the default Ubuntu Touch user [01:21] thanks [01:22] is possible dual os in n7 . android + ubuntu touch === lolcat is now known as lolant === lolant is now known as LoLcAt [02:30] sergiusens, the last pastebin i linked worked flawlessly that message appears to be a warning or something because /data was mouetnd and i could write to it and read it (thought it only had the lost+foudn folder) [02:40] annerajb: good [02:40] plars: if still around (and yeah, I know... :-P) https://code.launchpad.net/~sergiusens/phablet-tools/lp1211708/+merge/180031 [02:40] i just have to figure out why if i copy the rootfs into /data it breaks -_- [02:41] i been looking at pastebin online and apparently some devices work fine with that message so it may not be a failure. thought somebody is calling kernel_restart [02:41] Restarting system. [02:50] sergiusens: +1 from me on lp1211708 MP [02:57] cjohnston: great, if you approve I'll Happrove [02:58] Done [02:58] ty [02:58] I'm not special though :-) [02:58] no one should be :-) [02:58] heh [03:00] thanks for the review! [03:00] ah... Day changed... I'm going to bed! === Namidairo`bnc is now known as Namidairo [03:01] heh. g'nite [03:01] night! [03:37] sergiusens: mhall119: Could you please give me more details on what features the developer mode blocks? You could file bug too :) [03:52] bzoltan1: I think they're offline, it's late in their timezones [03:54] mfisch: I hope that they will see my answer when they wake up === chihchun_afk is now known as chihchun [03:57] bzoltan1: surely ;) [03:58] mfisch: at least I see them in the list of nicks who are here :) So it depends on their client... my pidging beeps when somebody mentions my name :) [04:44] Hi All [04:50] Need guide to install ubuntu-touch to my samsung tab 2 [05:23] Alo [05:23] Anybody? [05:29] RaziRazak: it isn't one of the supported devices. have you looked at the list of community devices? [05:31] Ooo [05:32] So p5100 are not able to be support by ubuntu-touch is it [05:33] https://wiki.ubuntu.com/Touch/Install#Supported_devices_and_codenames [05:33] Any future plan to for p5100? [05:34] no idea [05:34] Depends on whether someone wants to port to the p5100 [05:34] you can see if anyone from the community is working on it [05:35] RaziRazak: Oh, https://wiki.ubuntu.com/Touch/Devices says that someone's ported to the p5110. [05:39] P5110 is the model without gsm [05:40] P5100 is the model with gsm [05:45] Hi again [05:46] Disconnected just now === ivanka__ is now known as ivanka === arunprasadr_away is now known as arunprasadr [07:28] sil2100, ping [07:31] tvoss_: pong [07:32] good morning [07:36] Good day! :D [08:01] tvoss_: did anything happen? [08:02] sil2100, nope :) can you add ci and autolanding for lp:content-hub? [08:03] tvoss_: let me take a look [08:03] sil2100, thx, the packaging setup is rough to say the least ;) [08:04] tvoss_: will we be daily releasing it in the nearest future? [08:05] sil2100, might well be, yes [08:09] Does aybody know, p5100 can support ubuntu touch or not? === Namidairo is now known as Namidairo`bnc === Namidairo`bnc is now known as Namidairo === dholbach_ is now known as dholbach [08:58] Good morning all, happy Soyuz TM-25 landing day! [08:58] happy what day ? Hello to btw :D [09:03] didrocks, https://code.launchpad.net/~thomas-voss/location-service/upstart/+merge/180068 [09:03] sil2100: mind having a look? ^ [09:04] didrocks, tvoss_: looking [09:04] sil2100, great, thx === seb128_ is now known as seb128 [09:13] tvoss_: looks ok and builds ok, approved === alfonsojon is now known as alfonsojon|away [09:19] sil2100, thx [09:21] * tvoss_ wonders why the phone is taking 25% RAM on the phone [09:22] hey is there some way to sync ubuntu touch with windows pc to transfer some data on nexus 4? [09:24] N33Q, you can use adb push ... or install openssh-server on the phone and use scp/sftp [09:24] (teh latter via network indeed) [09:25] tvoss_, because its a phone ? [09:25] * ogra_ would prefer if the phone took 100% of ram on the phone :P [09:26] asac, new image is testing ... FYI [09:26] ogra_: i suspect we wont be happy though [09:26] well, its looking good so far [09:26] we have two runaway processes still ... unity8 and whoopsie :) [09:26] 40 of 40 tests green [09:26] yeah... lets hope for a bit of luck i guess [09:27] oh, whoopsie is still not fixed ? [09:27] not sure... didnt see people dancing, so i assumed not [09:27] https://bugs.launchpad.net/touch-preview-images/+bug/1211417 [09:27] Launchpad bug 1211417 in whoopsie (Ubuntu) "whoopsie takes 100% CPU on the phone" [Critical,Confirmed] [09:27] ev: news? :) [09:27] ogra_: what image is this with? [09:27] ev: can we disable it? [09:28] ev, no image, the tests just started and look all good ... but asac expects us to hit a 100% cpu issue again [09:28] or should we kick it out? [09:28] ogra_: well, both unity8 and whoopsie have known looping bugs that happen eventually. we have to be very lucky :) [09:28] well I'd prefer we see if it's still broken in the tests before we do something drastic like kicking it out [09:29] ++ [09:29] people suspect its the part the blocks our images from passsing tests [09:31] I've been hammering at it with valgrind and splint all morning. Found some minor niggles, but nothing that would point at memory corruption yet. I haven't been able to reproduce it with the latest phablet-flash cdimage-touch --pending yet, but if I can get it to appear, I'll try downgrading whoopsie to see if that resolves it [09:31] I'll have another look at the syslog code, as that's the only big change we've made in quite a while [09:33] still i would like to know how we can disable it or if the package has to go out in case we still have a non-green baseline today [09:34] in /etc/default/whoopsie i'd guess [09:34] or by overriding the upstart job from another package [09:34] ogra_: can we do that without uploading whoopsie itself? [09:34] the latter, yes [09:35] like in an override logic we do during image production? [09:35] ok [09:35] lets see what happens [09:35] echo manual >/etc/init/whoopsie.override [09:35] thats all thats needed [09:35] ogra_: curious, where would we put such code? [09:36] in another package? or in some image production scripts? [09:36] yeah, disabling the job is probably preferable as it wont mess with user's configuration [09:36] well, since lxc-android-config carries many such things (though all container related) i would put it there too ... since it is definitely temporary anyway [09:36] ok, but that requires a full package build etc. [09:37] but we can also do it at image production from livecd-rootfs [09:37] everything does [09:38] all bits that can make changes at that level of an image build are in packages [09:38] so an upload is unavlidable ... . [09:38] *unavoidable [09:40] ogra_: so image production logic is in packages itself? [09:40] no bzr branch ? :) [09:40] sure, there are bzr upstream braches for the packages :) [09:41] hehe [09:41] you know what i meant [09:41] bur yeah, rootfs creation is all in packages [09:41] so no [09:41] kk [09:41] image publishing and the whole build management as well as pre and post mangling are in cdimage and not packaged [09:41] but the build itself uses packaged bits [09:46] asac, did cypher and chicken talk to you about the bluetooth and indicator changes ? [09:46] didrocks, you can subscribe to to be notified of revisions. The downside is that you'll be notified of revisions to PC-only parts as well. [09:46] mpt: do you update it so often? [09:46] (like should I be scared about the spam? ;) [09:48] didrocks, ... it varies. 24 times this month, once last month, not at all in June. [09:48] mpt: ok, so should be fine, subscribing === gusch is now known as gusch|lunch [10:01] ogra_, does scripts/touch has some sort of crc?? [10:02] if i modify a line on it. it breaks. at first i thought it was space but i could copy the whole file and it woudnt break as long as i didnt modify the touch script. [10:02] weird part is that if i modify the touch script. and it's breaking it "works" as soon as i format /data [10:07] morning all [10:07] no, there is no hashsum or anything [10:09] ogra_, well then something really mess up is hiding in my initrd lol. I thought it was permission when repacking it since i never see any of the prinouts of my touch script when it fails just the watchdog. [10:11] well, it could be that your boot.img gets to big for the partition [10:11] have you checked that ? [10:11] (then it might go corrupt on flashing) [10:12] ogra_, i thought about it that's why i copied the touch script with another name to the boot.img root and it worked fine [10:13] so i have two touch image files. I also removed lines from the touch/script to see if the smaller size would help (removed the stuff that dosnt work on my cellphone like the find dev) [10:16] btw, did you try putting path=/dev/.... on your kernel cmdline ? [10:16] and not editing the script [10:17] wasnt that not implemented yet and sergiusens was going to work on it? [10:18] no, i'm going to work on it, but it might already work [10:18] iirc the kernel exports all cmdline args to the environment ... [10:18] i'm not sure if it gets handed through to the tough script though [10:19] but i think it would be owrth to try [10:19] (though that doesnt fix the sleep issue i guess) [10:20] ogra_, yeah i can try it it still makes me want to know why is this happening when i modify the script because if i made the script smaller same thing happened and that rules out space which was the only thing worrying me. [10:20] anyway how do you pass the cmdline args?? [10:21] abootimg -i /path/to/boot.img | grep cmdline [10:21] first check if there is an existing one [10:21] then: [10:22] abootimg -u /path/to/boot.img -c "cmdline= " [10:22] this is a samsung device abootimg dosnt work [10:22] boot.img: no Android Magic Value [10:22] boot.img: not a valid Android Boot Image. [10:26] oh, then you need to look in your BoardConfig.mk i think [10:27] BOARD_KERNEL_CMDLINE := console=ttyFIQ0,115200 init=/init no_console_suspend [10:27] is that no_console_suspend desired? [10:27] i read on the flipped page about console having to be on or something [10:30] nah, thats fine [10:30] so just add a path= there [10:30] wow, that init=/init is evil ... what a luck ubuntu initrd uses /init by default [10:31] (usually the kernel would iterate over several locations, that option forces it to only accept /init) === alan_g is now known as alan_g|tea [10:36] ogra_, all right i am building it again it's still using the heavily modified touch script thought. [10:37] still building cm-10.1... [10:38] my source for cm-10.1 builds fine in UT.. does not with the actual cm-10.1... how funny is that.. I have to cherry pick a patch [10:39] for tinyalsa.. I guess that is why [10:41] asac, ogra_: what's the jenkins job that was failing due to whoopsie hitting 100% CPU? I'd like to keep an eye on it, rather than waiting for people to come to me. [10:42] ev: we dont see it in jenkins yet' [10:42] oh right [10:42] I thought that's how all this started [10:42] ev: it came out of investigations because we saw flakiness... we are about to land a test that ensures the system calms down [10:42] that will reveal that [10:42] * ev nods [10:42] so folks see it locally by just rebooting mako [10:43] ev: the test that triggered us looking for sources of noise, is still failing [10:43] on mako [10:43] which might indicate that its still there [10:44] asac: I take it someone is digging to find out definitively? [10:44] ev: well, the looping whoopsie is a bug on its own [10:44] people are digging in the dark [10:44] looping? [10:44] the more we can rule out the better [10:44] I noticed you referred to it as such before [10:44] how is it looping? [10:45] ev: that it consumes 100% [10:45] oh that bit, right === alan_g|tea is now known as alan_g [10:45] asac, did the testers make sure it isnt upowerd that misbehaves as usual ... and whoopsie only being fallout of that ? [10:46] sure, I'm just not yet convinced that it's in whoopsie. valgrind didn't surface anything and I'm not able to as yet reproduce it on the latest proposed mako image [10:46] (i.e. are you sure whoopsie is the only process at 100%) [10:46] ev: but you could reproduce [10:46] asac: I was able to briefly reproduce it with the previous proposed image [10:46] sure thats enough [10:46] bug 1190792 has often misled us before [10:46] bug 1190792 in touch-preview-images "ueventd in a busy loop on container-flipped image" [Critical,In progress] https://launchpad.net/bugs/1190792 [10:47] ogra_: but we would see ueventd also going high on top [10:47] that wasnt observed from what i remember [10:47] ueventd is chewing away at 100% cpu on mine, for what it's worht [10:47] yes [10:47] not saying its not related [10:47] ev reboot [10:47] on my mako when it first showed a high whoopsie cpu, it also had a powerd crash in /var/crash.. [10:47] what is a problem are those looping things. thats a super grave bug [10:47] [10:47] if you think in the phone context [10:48] processing .crash files perhaps ? [10:48] and the fact that it makes our autopilot tests more flaki (due to UI tests being async with timeouts etc.), doesnt make it less severe :-P [10:48] popey: ok powerd crash can explain that [10:48] thanks [10:48] popey: can you still reproduce? [10:49] no, trying [10:49] ok good. do you see ueventd? [10:49] it didnt happen with todays flash [10:49] flashing at the moment [10:49] cool. [10:49] asac, everyone sees ueventd every other boot [10:50] guess that will be visible with the settle test then as well :) [10:50] i dont see it on maguro [10:50] yeah, i have a machine with ueventd at 100% now [10:50] but no whoopsie [10:50] we have reduced the racyness as much as we could, buut the real fix waits for an upstart upload [10:50] for me its unity8 keeping my thing hot :) [10:51] asac, eaxctly, maguro is a lot slower [10:51] ogra_: where is that upstart upload? [10:51] so we have a fix? [10:51] so it doesnt see the race trhat often [10:51] ogra_: ueventd goes crazy right after boot? [10:51] asac, the fix is an upstart bridge into the container ... jodh is at debconf so i doubt we'll see it landing this week [10:52] ueventd goes crazy if udev tries to do something with devices while ueventd hasnt finished yet [10:52] we need the equivalent of "udevadm settle" for ueventd [10:52] ogra_: its a start race condition? [10:53] which the upstart bridge is supposed to provide [10:53] yes [10:53] ok, so we will see ueventd rumbling right after boot.. thats good [10:53] only happens on startup and more often on mako [10:53] but it persists until reboot [10:53] righ [10:53] t [10:54] thats good. adding my test to default will then show runs clearly that suffer from this effect [10:54] right [10:54] and we can retry the test until we hit a boot where we dont see that before we can take any results for serious [10:55] well, dont the tests all reboot the device before starting ? [10:55] or does that only happen once ? [10:55] * ogra_ thought they use reboot to gain a clean state [10:55] so you would have to hook into each tests separately if thats the case [10:55] *each test [10:56] so i just clean flashed and I have a powerd crashdump in /var/crash [10:56] yeah, powerd needs to wait for the android event [10:56] * ev lunches [10:57] it accesses the sensorservice ... which it only can after the container is up === MacSlow is now known as MacSlow|lunch [10:57] ogra_: no ... we run multiple tests in one boot sequence... the reboots are not nicely marked in the dashboard [10:57] (currently it comes up with dbus ... thats to early, so it crashes until the proximity sensor is there) [10:58] i want them to run the default test suite after each boot [10:58] and dont continue if that fails [10:58] wow, you get quite a few apparmor_parser processes just after boot.. is that one per app? [10:58] rather reboot [10:58] ogra_, the command line parameter was never passed off [10:58] # cat /proc/cmdline [10:58] console=ttyFIQ0 no_console_suspend=1 androidboot.serialno=39 [10:58] popey, i think it processes the policy files once on first boot [10:59] annerajb, check your kernel config, probably it comes from there instead ... [10:59] ok [10:59] ogra_: whats the process for working around the ueventd issue? kill it? [10:59] ogra_: DEITY just send us an image from the future :) [10:59] asac, so my theory is ... powerd dies (on purpose atm) ... produces a crash file ... whoopsie sees the file and processes it ... which takes 100% cpu [11:00] 20130816 touch_ro mako [11:00] haha [11:00] maybe that will show us what we need to change in order to fix it? [11:00] :) [11:00] ogra_: I've had plenty of powerd crashes on my mako [11:00] without it chewing away at 100% cpu [11:00] hard to compare ... we need to disassemble it :) [11:00] basically one every boot :) [11:00] right, lunch for reals [11:00] ogra_: yeah. thats fine if its that way [11:00] ev, well, that at least proves my theory of powerd dieing on every boot :) [11:01] but we could in testing maybe disable whoopsie and run it explicitely in a more controlled manner at the end [11:01] ev: i guess there is a way to stop whoopsie and then run it explicitely once to scrape all .crash files? [11:03] ogra_: so that image is scary ... where is a clock offset? [11:03] on our image builders? i guess it uses what is in this media-info thing [11:03] dunno, probably on system-image.ubuntu.com ? [11:04] media-info isnt touched on system-image images [11:04] but it provides its own additional stamp on top [11:04] yeah whatever that is then [11:04] (now dont ask me where :P) [11:08] ogra_: so flakiness confirmed :) [11:08] http://reports.qa.ubuntu.com/smokeng/saucy/image/3512/ [11:08] Saviq: ^^ :-P [11:09] asac, told 'ya! :P [11:09] ogra_: is there anything else we could do to avoid this ueventd issue until it lands? [11:09] asac, although it might be that *something* made it worse, for sure [11:09] Saviq: yeah no more annoyance for at least 16h :) === schwuk is now known as schwuk_away [11:09] Saviq: i am confident that mterry can nail it down if we dont let it go :) [11:09] https://code.launchpad.net/~ogra/powerd/fix-semsor-race/+merge/180095 [11:10] at least give us confidence on what is going on [11:10] yeah [11:10] asac, no, we cant do much more for ueventd [11:10] ogra_: nothing hacky wacky? [11:10] i need to sort out some dependencies for the ubuntu-touch package. It depends on chewie (netowork + sound indicator backends). We need to remove this dependency, but ensure that chewie is still installed on phone for now. [11:10] given we have understood the problem and the real fix is coming its a valid approach imo [11:10] you can add even more sleeps but i doubt that will fix much [11:11] (and it adds to the boot time indeed) [11:13] i think we need to move chewie from core see to shell seed ? [11:13] sforshee, mfisch, can one of you approve https://code.launchpad.net/~ogra/powerd/fix-semsor-race/+merge/180095 [11:13] or something... [11:14] dednick, we dont have such a differentiation in the seeds [11:14] there is just touch and adk [11:14] *sdk [11:14] ogra_: but there is diff between recommend and depends? === schwuk_away is now known as schwuk [11:15] dednick, no, we fully suppress recommends ... whats not seeded or a hard dep isnt in [11:16] * ogra_ hopes we can clean that up at least for 14.04 [11:16] ogra_, the kernel had the argument in the config [11:17] sigh, do there's no way we can remove chewie [11:17] dednick, i think the plan is to remove it soon, but the indicator porting work isnt fully done as i understand [11:17] annerajb, so try changing it there then :) [11:18] ogra_: yeah, i was hoping we could remove it as a dependency, indicator-network removes chewie... bit of a ballsup. [11:18] yeah already flashing it [11:18] ogra_: nevermind. will just sort out dep of network then [11:18] dednick, https://code.launchpad.net/~mathieu-tl/ubuntu-seeds/ubuntu-touch.chewie/+merge/179937 [11:19] dednick, but that will cause regressions until all indicator work is done [11:19] ogra_: ah, ok. thanks. [11:19] so we either needs approval from asac to allow such a regression (and someone t explain what actually regresses) or we need to wait === gusch|lunch is now known as gusch [11:21] my understanding is that indicators will start landing later this week one by one [11:21] ogra_: no, it's ok. i just wanted the ability to install indicator-network for testing, but it conflicts with chewie, so it makes life a bit hard. [11:21] i have no idea what this chewie thing would fix us [11:21] dednick: can you remove it locally? [11:21] asac, chewie is the current backend for the current indicators [11:21] asac: with a bit of effort [11:30] ogra_, it didnt work even thought cat /proc/cmdline shows it there it dosnt seem to get pass to the touch script (or it get's overwritten [11:31] right [11:31] i'll try to add a cmdline parser to the script later today [11:31] to actually fix that open bug [11:32] ogra_, thanks [11:32] ima head to work later === alan_g is now known as alan_g|lunch === MacSlow|lunch is now known as MacSlow [12:09] tvoss_: hmm, do you know why suddenly something from the platform stack wants to pull in libgles2-mesa on the desktop? [12:09] tvoss_: I mean, do you know of any change that might have resulted with such a change? [12:09] sil2100, nope, would need to investigate, too [12:20] ogra_, let me know when you have the merge so i can pull the diff from it :D [12:26] sil2100: you have the reason for the "something" [12:27] /var/log/upstart/otto-setup.log: Installing libgles2-mesa as Depends of libmirplatform [12:28] sil2100: I think we don't have libmirplatform installed as well on the desktop, so need to refresh the dependency list ^ [12:28] didrocks: oh, indeed, I didn't know we had such a nifty logging in otto! [12:28] I mean, forgot about that [12:29] didrocks: libmirplatform is in the list already btw. [12:31] didrocks: https://code.launchpad.net/~sil2100/cupstream2distro-config/add_libmirplatform_extra/+merge/180120 [12:32] didrocks: as for HUD - the manual publishing needs to be held, since the changelog entry in that packaging diff looked wrong, so I edited that [12:32] didrocks: also, will have to see if we won't have to release unity-lens-apps in the same time with the new libcolumbus bindings [12:37] sil2100: approved [12:37] didrocks: thanks [12:37] yw === alan_g|lunch is now known as alan_g [12:56] sil2100: can you add this to your list for daily release? https://code.launchpad.net/~sergiusens/cupstream2distro-config/unity-mir/+merge/179983 ricmm should be able to answer questions, by the name you can guess what it's for ;-) [12:56] sil2100: greyback as well [12:56] fginther: reping on ^^ [12:57] sergiusens: ack [12:57] sergiusens: we were just discussing that in -desktop ;) [12:57] sergiusens: perfect timing [12:57] sergiusens: so, could you make one small modification there? [12:58] sil2100: as many as you want [12:58] sergiusens: ah! Scratch that, it's fine as it is ;) [12:59] sil2100: my only doubt was if I should put it in the mir or unity stack, but since it's glue code for unity to talk to mir, the unity stack felt like the right place to be [12:59] sergiusens: I'll poke now Francis to prepare all the CI bits [12:59] sergiusens: I had the same thing, but I personally wanted it in platform [13:00] sergiusens: but unity8 seems like the right place - the unity8 guys are upstream for it, so I guess it's more sane this way [13:00] sil2100: yeah, I had that one in mind too, but then mir should also be in platform (personal opinion) [13:00] Makes sense [13:02] seb128, ack [13:05] hum [13:05] the thing is that the API of Mir is unstable [13:05] not sure if it's using the server or client one [13:05] but the server one is ABI unstable [13:06] I hope the unity-mir API is stable at least? [13:07] we've passed $10M! :) [13:07] ++ [13:07] sergiusens, one comment [13:07] only 22 to go [13:08] sergiusens: ^ [13:09] * sergiusens fixes [13:09] fginther: should that be the default hook? [13:09] fginther: I will remove the hook line completely [13:09] sergiusens, that would work [13:11] fginther: hmmm, check license headers is not in the default hook, want me to set that? === boiko_ is now known as boiko [13:11] sergiusens, yes, that works for this stack [13:11] fginther: if projects don't comply, they should start fixing it === _salem is now known as salem_ [13:13] sergiusens: your branch got merged btw ;/ [13:13] sergiusens: so you'll have to fix it in a separate one [13:13] sil2100: how did that happen? [13:13] oh, ic [13:13] no worries [13:14] sergiusens, argh I must have not flipped the state right [13:14] sounds like some american lawyer trick [13:14] "flipping the state right" [13:15] just requires the right amount of $$ [13:15] heh [13:15] ogra_: my chromebook is on precise again, blazing fast, rock solid! [13:15] sergiusens, haha, raring isnt bad either :) [13:15] ogra_: but I get fast unity2d :-) [13:16] sergiusens, we just need to get the Mir guys to make Mir work without the android crap and should get XMir o saucy [13:16] *on [13:16] ogra_: with which drivers are you aiming for? :-) [13:17] mali [13:17] desiring would be a better word [13:17] fginther: https://code.launchpad.net/~sergiusens/cupstream2distro-config/unity-mir/+merge/180129 [13:17] the ones i use with the compiz unity on raring atm [13:18] * ogra_ uses the manta drivers on his chromebook [13:18] ogra_: which would be similar to building this: https://github.com/hrw/chromebook-mali-driver [13:18] oh, the manta ones, I have no idea where the ones in that git repo come from [13:18] oh, they are the google ones [13:18] i think copied out of the chromeos rootfs [13:19] ogra_: from http://commondatastorage.googleapis.com/chromeos-localmirror/distfiles/mali-drivers-0.45-r96.run [13:19] anyway, the mantra drivers work fine [13:19] *manta [13:19] * sergiusens needs to try that again [13:20] but according to tvoss_ there is no way to do gralloc stuff without hybris [13:20] and there is no android port to the chromebook that i know of [13:20] mzanetti: ping [13:21] sergiusens, thanks [13:21] tmoenicke_: pong [13:23] fginther: in the meantime... [13:25] fginther: we added to the platform stack a new project, content-hud, which would need CI and merger prepared [13:25] fginther: it's daily_release: False now, but will be released in the nearest future [13:26] sil2100, ack, when sergio's merge is finished, I'll deploy all the updates === chihchun is now known as chihchun_afk [13:27] fginther: excellent :) THanks! [13:31] ogra_, except for if you have drivers implementing the android HAL (i.e., the interfaces in gralloc.h) compiled against glibc [13:31] well, i dont ... [13:31] balloons: weather app appears to still be failing bad, but getting 1 test passed at least [13:32] but the android mali drivers work with compiz here [13:32] without issues [13:32] ogra_: you can always fall back to the chromebook ones just in case... [13:32] yeah [13:32] point should be moot thee [13:32] *there [13:32] but i doubt they use gralloc [13:37] annerajb, http://paste.ubuntu.com/5985004/ [13:42] ogra_, thx ill give it a try later [13:42] annerajb: do you still need that sleep? [13:43] ogra_: what the...... http://reports.qa.ubuntu.com/smokeng/ you're making builds 2 days in the future now what???? [13:43] sergiusens, looks like his flash controller operates slow on start [13:44] sergiusens, i probably do still need it [13:44] davmor2, yeah, asac wanted to be able to backport the fixes from friday so we decided to build fridays image in advance [13:44] but talking to ogra_ yesterday udevdam settle should handle that [13:45] ogra_: retroactive bug fixing? [13:46] sergiusens, pro-retroactive bugfixing indeed :) [13:46] sergiusens, and ogra_ :) [13:46] * davmor2 begins to wonder if asac and ogra_ are the new Doctor and his companion with their timey wimey shenanigans [13:46] looks like ubuntu-touch will be coming to xperia z soon ;) [13:46] bunching 1 hour with no errors yet [13:47] ikillcypher, :D [13:47] congrats [13:47] thanks to mamenyaka [13:47] davmor2, but honestly, that looks like brakage in stgraber's playground [13:54] ogra_: cool, the sensor is emitting the event now I guess. Thanks for fixing that [13:54] FAILED :( [13:54] OH NO [13:54] cc1plus: all warnings being treated as errors [13:55] home/ikillcypher/xperiatouch/out/target/product/yuga/obj/KERNEL_OBJ/usr/include/video/msm_hdmi_modes.h:214:43: error: narrowing conversion of 'i' from 'int' to 'uint32_t {aka unsigned int}' inside { } is ill-formed in C++11 [-Werror=narrowing] [13:55] mfisch, np, i was looking for the reason the sensorservice races sometimes for weeks ... luckily sforshee poinyted out that powerd treis to access the proximity sensor [13:55] davmor2 sergiusens ogra_, is there a way to get the YYYYMMDD.X build number on an upgrade? essentially what you are seeing there with the build being in the future is that adb shell system-image-cli -b gives the 'version' number, and media-info has the original build number that was installed.. [13:55] We need a way to get the upgraded build (YYYYMMDD.X) build number instead of the version number [13:56] cjohnston, ah, yeah, system-image images use their own stamp ... you should read that instead and use media-info for the initial install media info [13:56] now dont ask me where that stamp lives :P [13:57] ogra_: this is checking though to see if the install was upgraded.. So if you started on 20130808 and upgraded to today's build. media-info would still show 20130808 [13:57] yes [13:58] media-info never changes [13:58] but there is a separate stamp file the system-image images use [13:58] ogra_: any idea if we're still seeing whoopsie at 100% CPU in the lab, even with the latest test? [13:58] ogra_, [13:58] http://pastie.org/private/3rydde6mpycywz54rwchg [13:58] any idea how to fix this [13:58] ev, nope, ask someone who is near the lab ... i.e. plars [13:58] plars: oh hai [13:59] ev: we were yesterday [13:59] not sure about today [13:59] cjohnston: yesterday was a very different day :) [13:59] I saw it with yesterday's bits too, just not today's [13:59] ev: I am nowhere near the lab, but yes we are [13:59] :'( my errors guys ?? [14:00] plars: who can I talk to for more information? [14:00] ev: me [14:00] :) [14:00] ev: though I'm not sure I have too much more than you already have, I'm a bit confused that you are not able to reproduce it as easily [14:00] ev: for the rest of us, it's hit or miss, but rebooting the phone once or twice should see it for sure [14:01] ev: I'm flashing the latest image on my phone locally [14:01] plars: phablet-flash cdimage-touch --pending ? [14:01] ev: yes [14:02] ikillcypher, how often did you try googling it before asking us ? :) ... this is all device specific failures ... either related to the compiler defaults (-Werror) or caused by a config option in your kernel [14:02] ok [14:02] :( === chihchun_afk is now known as chihchun [14:03] ev: we also have https://bugs.launchpad.net/touch-preview-images/+bug/1190792 that describes eventd doing the same thing to us at time [14:03] s [14:03] Launchpad bug 1190792 in touch-preview-images "ueventd in a busy loop on container-flipped image" [Critical,In progress] [14:03] plars: is there anything special you're doing after flashing? Is wifi on? [14:03] ikillcypher, try to do some research yourself first :) [14:03] plars, are we 100% sure the whoopsie thing isnt just fallout of ueventd ? [14:03] ev: yes, it's flashing, then setting up wifi right away [14:04] ogra_: no idea, but that's why I mentioned it to ev just now [14:04] bzoltan: what do you mean "what features the developer mode blocks"? [14:04] ogra_: just trying to find the source to have a look :) [14:04] do you mean what's blocked when developer mode isn't enabled? [14:05] compiling cm-10.1 with 3.0.8.. the real cyanogenmod with android.. mainly to verify I get a screen. Then try again with ubuntu touch [14:05] Anyone know a current kernel config for say, mako with ubuntu touch [14:06] I looked at phablet git sources, but they all had _NZ set for not used [14:06] ev, of ueventd ? [14:06] ev, thats on phablet.ubuntu.com in the gigantic git tree ... or in the android package in saucy [14:07] whoop [14:07] I brunched mako, and the kernel folder is not there.. prebuild kernel? [14:07] mhall119: yes [14:08] mhall119: I am going to cut the developer mode to pieces: 1. application development mode 2. platform development mode [14:08] application developer mode is the same as normal user mode, AFAIK [14:08] nevermind, found it [14:08] barry: if you have any question before my EOD on the wiki rewrite, do not hesitate :) [14:09] didrocks: thanks! and great meeting [14:09] barry: I hope you won't need any medecine in rereading the etherpad ;) [14:11] :) [14:11] mhall119: except it needs ssh access [14:12] ah, right, and openssh isn't installed by default, and not installable unless you're in platform developer mode [14:12] bzoltan: can QtCreator's plugin use adb shell instead of ssh? [14:14] mhall119, that wont be open forever either [14:15] kernel requirements for lxc The requirement of kernel config is listing in http://lxc.teegra.net/ [14:15] i would say if you devellop on/with the device ... we expect the device to be in developer mode [14:15] hmmm LXC has been in the mainstream kernel since 2.6.29 but there are several configuration options that need to be set so that it can be used. [14:16] app armor was put in around 2.6.36 or so, seccomp should be enabled, and cgroups of course [14:16] good info [14:16] fanotify the same, I was told 2.6.36 is basically a test of new features that are put into the beginning of 3.0 kernel [14:16] mhall119, I generally use the command line to avoid the full developer download, bzoltan going to fix it [14:18] AppArmor, seccomp and cgroup confinement consistently to all apps... http://loco.ubuntu.com/planet/592-may-2013-ubuntu-developer-summit-summary/ [14:18] haix [14:18] changed int to uint32_t === alan_g is now known as alan_g|tea [14:24] bom dia === alan_g|tea is now known as alan_g [14:37] 50 [14:37] bah [14:37] 60? [14:38] ogra_, reboot loop now [14:38] 69 ? [14:38] annerajb, great [14:38] annerajb, can you get into recovery without powering down ? [14:38] how? [14:38] so that you can check /proc/last_kmsg [14:39] holding the right buttons to get into recovery [14:39] while it reboots [14:39] the same way you usually get into recovery? [14:39] yeah [14:39] just make sure you dont powerd down ... [14:39] else the ram console gets flushed [14:40] (and /proc/last_kmsg would be empty) [14:40] hmm cant seen to get into recovery -_- [14:40] annerajb, argh ... there is a typo in the snippet [14:40] *angry fist* === arunprasadr is now known as arunprasadr_away [14:42] annerajb, there is a ;; missing [14:43] between line 30 and 31 in the paste [14:45] all right building another boot.img [14:47] ogra_, i seem to be having trouble getting into recovery now any ideas why? [14:48] nope [14:52] ChickenCutlass, do I need to enter a bug on the volume control? [14:52] pmcgowan, I guess. it works from command line. Not sure why it broke [14:52] pmcgowan, need the new indicators anyway [14:53] ChickenCutlass, strange, what handles the button? [14:53] pmcgowan, I beleive the server -- chewie or something [14:53] ChickenCutlass, ok so same issue, if sound indicator is landing can wait for that [14:53] pmcgowan, right [14:55] mhall119: for sure it can... but keep in mind that adb needs physically connected device ... and my plan is to give development access via usb/wlan too [14:57] bzoltan: you will need to do that thinking abuot image based upgrades and read only filesystems [14:57] and the fact that adb root will go away, as ogra_ mentioned [15:10] bzoltan: is the .desktop fix for the QtCreator click package creation in saucy? [15:11] mhall119, o/ [15:11] also, wow this is a lot of people here [15:11] beuno: bzoltan: so we need the QtCreator click creation fixed before we get people using the new MyApps [15:11] beuno, But now that you're here, it is complete. [15:12] that's like twice as scary [15:12] beuno: welcome to the high traffic channel [15:12] almost as much as #ubuntu :-P [15:12] * beuno already regretted joining [15:13] is anyone else having trouble with phablet-test-run? It's not running things on my manta [15:13] mhall119, hi, just to tell you about it, i've been working these days in these 2 apps: http://youtu.be/uYjG8Iq1cD0 - http://youtu.be/65vvhFt57DU - http://youtu.be/F-_Hsl1Ts14 [15:13] balloons: other than it locking uip mid-test...? [15:13] popey, no it connects and says runs 0 tests [15:14] oh wrong channel [15:14] gatox: video unavailable [15:14] weird [15:14] returns ok and ends.. it won't actually run ;-( [15:14] popey, maybe it's not accesible just yet [15:15] gatox is like the worst vacation taker ever [15:15] beuno, jejeejjee [15:19] sergiusens, [15:19] Oroku_Saki, [15:19] Package Complete: /home/ikillcypher/xperiatouch/out/target/product/yuga/cm-10.1-20130814-UNOFFICIAL-yuga.zip [15:20] xD Ubuntu-Touch On Yuga [15:20] ikillcypher, congrats !! [15:20] how did you get around the error in the end ? [15:20] like finally mamenyaka helped me [15:20] the guy who ported xperia ubuntu-touched [15:20] ah, i didnt see him around today [15:21] well google plus mate :D [15:21] :) [15:21] now I need to figure out how to back up my cm10.1 [15:21] and flash this o.o and get it working [15:21] yeah, you only made the very first step [15:21] there is still a lot ahead for a fully working port [15:22] well atleast getting it boot was something everyone wanted to see [15:22] fully working port isnt any where near maybe not now but that should be too hard at all [15:22] ogra_, i had to flash to CWM recovery instead of the ubuntu one to be able to get into recovery [15:22] got a dmesg [15:23] ogra_, Kernel panic - not syncing: Attempted to kill init! [15:23] annerajb, awesome [15:23] so it mounted [15:23] can you pastebin ? [15:24] abd shell cat /proc/last_kmsg | pastebinit [15:26] balloons / sergiusens is there any good reason why, when we run phablet-test-run it calls autopilot like this:- [15:26] exec_with_ssh autopilot run -o /tmp/test_results.xml -f xml $TESTSUITE [15:26] with -f xml, so the resultant log file is pretty much no use? [15:26] (for debugging where the test failures are) [15:26] what consumes those xml files? [15:26] sure they are, just usa an xml editor :P [15:27] popey: hmm.... jenkins [15:27] popey: just don't call it with xml generation [15:27] the xml is truncated, shows no detail of where failures occur, that I've seen [15:27] whereas a full autopilot log is actually useful [15:27] sergiusens, but then you cant read the log in an xml editor ! [15:28] popey: iirc, don't use -o [15:28] but I need -o so it spits a log out that i can collect [15:28] * popey fiddles [15:29] popey: so what you mention as a problem may be an autopilot issue [15:29] beuno: where can I upload a new package? [15:29] mhall119, I don't know, lets find out. Can you now not go back to the app and upload a new version? [15:30] beuno: I'm on https://myapps.developer.ubuntu.com/dev/click-apps/3/ [15:30] do I have to start all over? [15:30] ogra_, i dont think it mounted since i dont see the initrd print that says mounting [15:31] mhall119, I hope not. Is there a link next to "version number" to add a new version? [15:32] beuno: there's no link next to version number itself, but there is an "Edit" link to the side of each section, including the "Your App" section which contains the version number [15:32] annerajb, well, give me a pastebin :) [15:32] ogra_, http://paste.ubuntu.com/5985400/ [15:32] mhall119, ah, that's not very intuitive [15:32] beuno: a giant "Upload new Package" button would be great [15:33] clicking the edit button brings me back to the first step in the process too [15:33] hm [15:33] one sec [15:33] clicking on "Uploading your app" gives me a screen with: "Current version already uploaded. Before a new upload, you must update your application's version." [15:33] annerajb, hmm, i agree, i guess it is the sleep [15:33] oh, so I need to actually bump it from 0.1 to 0.2 on the first screen, and only then can I upload a new package? [15:34] (or the nonexistence of the same) [15:34] mhall119, yeah, although we fixed that. I think it hasn't been deployed yet [15:34] ogra_, well it's not printing the kmsg so i dont think it executed the mount [15:34] yes [15:35] it still dies during the flash initialiyation [15:35] beuno: bzoltan: also, there's nowhere on the Packaging form in QtCreator to change the version number of the package, how are developers supposed to do that? [15:35] ogra_, so add a sleep somewhere in there? [15:36] mhall119, yeah, you should file a bug for that. You can edit it in the advanced tab [15:36] mhall119, so yes, first bump the version, then you can upload. That has been fixed and we're deploying the update later today [15:37] beuno: I assume I can use 0.1.1 [15:37] mhall119, yes [15:38] mhall119, it uses debian rules to check if it's newer [15:38] annerajb, yeah, right before the mount ... start with something radical ... sleep 10 or so ... [15:38] gah, updating the version number brings me back to the details page, so now I have to go back through everything again to upload a new package [15:39] mhall119, I think [15:39] that with the new version added [15:39] oh, wait, now there's an Upload link next to the 0.1.1 version number [15:39] you may now have a link to up.. [15:39] right [15:39] again, sorry about that, it's fixed just not yet deployed :) [15:39] and by fixed, I mean it's been like this for years for software center [15:39] and we are slowly improving on it [15:40] beuno: ok, uploaded the new package, but my app is still in Rejected state, shouldn't that reset when a new version is uploaded? === jhodapp is now known as jhodapp|lunch [15:41] mhall119, I think I should of "ask for information" instead of rejected :) [15:43] mhall119, looks great now! [15:43] aaand I broke something [15:43] :) [15:46] poor davmor2, we're breaking everything before he even gets a chance :) [15:46] mhall119, I messed up, shouldn't of rejected it [15:46] can it be un-rejected? [15:46] yes [15:46] mhall119: save me breaking it [15:47] mhall119, yeah, I need to fiddle with with some things first [15:47] ogra_, same thing but this time before the crash i saw system-udevd starting. [15:47] (ie it crashes before the moutn much earlier [15:48] mhall119, done, back to you now [15:49] ogra_, where is system-udevd started because i dont seem it on the scripts/touch being started there (so i dont think we are reaching the sleep ) [15:49] beuno: I hope you gave it a thorough review I don't trust that mhall119 as far as I can throw him ;) [15:50] davmor2, he's pretty skinny though [15:52] * davmor2 wonders if I can break mhall119 code again it's been a while.......now where did I put that big hammer....... [15:52] davmor2: I hand-coded XML-RPC data, I'm sure nothing could possibly go wrong [15:53] published! [15:53] beuno: now where can I get a public URL to promote it? [15:53] mhall119: Yay! [15:55] ogra_: during the android guts on a flash I'm seeing a lot of text appear that didn't before should I be worried about that, I'll try and jot down the error this time :) [15:55] m-b-o, so weather had some failures again.. [15:55] davmor2, i always have text here [15:56] balloons: yes, don'know what happened... wokring on device and locally [15:56] ogra_: I have the last 3 flashes but didn't up until then [15:56] m-b-o, well there is some new improved ideas we could try for timing [15:57] hello everybody...may i ask some simple Qs here!? anybody could help with new tablet ubuntu... [15:57] ogra_: E:Can't open /cache/recovery/ubuntu_command seems to trigger the text [15:57] I'm wondering why the load avg is so high on my phablet, its not doing anything other than charging, I am logged in via ssh but this is the ouptu from top [15:57] top - 15:56:32 up 19:41, 2 users, load average: 7.74, 4.62, 4.16 [15:59] phablet-chris: paste the rest of top to pastebin? [16:02] sure thing http://pastebin.com/Ljgkeqxu [16:02] i just noticed though after I uplugged it the load avg dropped to 1.6 and is decrease [16:03] after I unplugged it [16:03] http://pastebin.com/4HCAZH4g [16:09] that was wierd, after uplugging it went back to normal, so I plugged back in and its still fine. I will keep an eye on it and post to pastebin if I see an anomally === alan_g is now known as alan_g|afk === alan_g|afk is now known as alan_g [16:12] mhall119, a public URL to your app? [16:12] probably becuase I was trying to play led zepplin 3 on you tube and it was stuck trying to dowlaod :) [16:13] dpm: mhall119 pmcgowan added "System apps" to https://wiki.ubuntu.com/Touch/CoreApps/Contents so the non-core "Core" apps (which I called "System apps" - got a better term?) show up on all core apps pages. [16:14] thanks popey [16:14] don't think I missed any out.. feel free to let me know (or add) if I did [16:15] can i install ubuntu phone on any android device??????????? [16:16] and a link to the download page plz... [16:17] can i install ubuntu phone on any android device??????????? [16:17] Pandix, https://wiki.ubuntu.com/Touch/Devices [16:17] can i install ubuntu phone on any android device??????????? [16:18] tnx w-flo ! === Namidairo is now known as Namidairo`bnc [16:19] is there a documentaion page for touch? I'd like to contribute as I use it [16:22] phablet-chris: we have a wiki, anyone can edit that [16:22] https://wiki.ubuntu.com/Touch is the main page, with other pages under it [16:22] if you have suggestions for new pages, just shout or edit away! [16:22] thanks [16:30] Are we planning on using plymouth at all in Touch? [16:33] mterry, we would either need a new framebuffer driver in plymouth or enable fbcon on all devices (which isnt possible for most ports due to kernel breakage that will occur with the GL driver then) [16:33] (mali doesnt allow fbcon next to it for example) [16:33] ogra_, do you happen to know offhand the replacement for plymouth-ready in upstart jobs? [16:35] mterry, no, currently there is a diversion/override for all plymouth bits in the touch images [16:35] ogra_, OK, thanks [16:35] i imagine we will want some kind of boot splash, and it would be good if that could be plymouth [16:36] but i guess the only way would be some kind of new framebuffer driver then [16:36] inside plymouth [16:36] well, I reflashed to stable to get my phablet-test-run working.. heh, now it only gives me an ssh error :-( ssh_exchange_identification: read: Connection reset by peer [16:36] balloons, you might need to run ./run_on_device -s [16:36] bad peer ... this evil norwegian guy [16:36] balloons, from unity8 trunk [16:37] heres another spike, now I know for sure nothing running other than top. and its unplugged http://pastebin.com/vfKZ7P3h [16:37] mterry, ty, I'll try that [16:37] w-flo, i wanna thank you for your work on the vision it has helped me on the epic a lot :D [16:37] balloons, (it sets up ssh and ppa stuff on the device) [16:37] ogra_, i miss any message if you ever sent me one after my last post forgot i was here and closed the laptop and went to get food [16:38] lucky you then [16:38] * ogra_ didnt have food yet :) [16:38] annerajb, so the sleep didnt help ? i wonder if we just see log spam that prevents us from seeing important stuff [16:39] w-flo, you had any trouble on the vision while booting from the initrd? (apart from having to hardcode the path) [16:39] ogra_, i dont think it has gotten to the sleep before it crashes [16:39] thats weird [16:39] like the line before the kernel panic is starting system-ueved [16:43] ogra_, systemd-udevd[154]: starting version 204 [16:43] that's the last line before the kpanic [16:44] yeah [16:44] but you got into adb before [16:44] so i dont think it crashes out of the touch script [16:44] i can adb in recovery but not on the loop [16:44] unless you made any heavy kernel changes [16:45] i reverted all of them yesterday [16:45] you could adb when it didnt mount [16:45] oh yeah when it didnt mount or when it did and /data was empty it worked [16:45] into the busybox shell ... or couldnt you ? [16:45] right [16:45] so i'm pretty sure the logging is out of order thanks to the flash controller [16:46] * annerajb glares at samsung [16:47] are you sure you fixed the typo properly ? [16:48] case ${x} in [16:48] datapart=*);; [16:50] was that the right line? ogra_ [16:51] OMFG [16:51] ikillcypher, it worked? [16:51] SUCCESSFULLY PORTED UBUTU-TOUCHED TO XPERIA Z [16:51] HELL YES [16:51] ikillcypher, lol [16:52] IT BEAUTIFUL MY EYES TEARS [16:52] your baby [16:52] is ubuntu-touch supported in any form on amd64? the ubuntu-touch package was installable from the default saucy repos after installing from saucy daily desktop ISO. i'm using acer iconia w500 x86 tablet, but after the first boot after installing the ubuntu-touch package and dependencies, it'll just hang on boot [16:52] ikillcypher, YAY [16:53] wtf there isnt any back key [16:53] uh no back key now Im stuck here o.o in events [16:53] ikillcypher, you have to learn to use the ui [16:53] tj____, no, it isnt being tested on x86 at all currently ... [16:53] ikillcypher, you swip from left to right across the whole screen [16:53] swipe* [16:53] tj____, (the plan is to support x86 in the future indeed) [16:54] ogra_, ok thanks [16:55] ikillcypher: nice [16:55] hardwarewise the old w500 is finally supported by default ubuntu kernel with saucy, which was nice. i bought it a few years ago and something's always been broken with it (touch screen, bluetooth) [16:55] would be nice to finally put it in use with ubuntu touch ;) maybe i try stock saucy + e17 for now [16:56] well, if Mir runs you should be able to use unity8 and Mir [16:56] it is abit of an issue [16:56] just not under the ubuntu-touch rask [16:56] looks like unstable :S [16:56] *task [16:57] the ubuntu-touch meta package pulled unity8 stuff but afaik no Mir [16:57] ikillcypher, now you have to start optimizing it :D [16:57] yeah, Mir on touch is in progress [16:57] should land soon [16:57] the point is that the ubuntu-touch meta is currently knitted with an android container in mind [16:57] w500 has an amd APU that works with the KMS radeon driver [16:57] well Im going to flash back cm o.o since this is unstable [16:58] we need to decouple these two parts fro palin x86 installs to work [16:58] ikillcypher, remember to post on xda so other people can join in and optimize it [16:58] *for plain [16:58] I will [16:58] ogra_, ok i see thanks for the info [16:58] it will happen before release ... until then just using the parts by themselves should work [16:58] how do I screenshot o.o [16:59] ikillcypher, i think the release notes link from the topic covers screenshots [17:00] whats with my setting turning black === Namidairo`bnc is now known as Namidairo [17:01] ikillcypher, your setting ? [17:02] it is highly buggy on my phone [17:02] open settings makes it black [17:03] ikillcypher, so debug it :) [17:03] well open browser fixes it [17:03] lol [17:03] dude how do I take screenshot === jhodapp|lunch is now known as jhodapp [17:04] read the change log ikillcypher [17:04] what changelog [17:04] im new o.o remember [17:05] ogra_, is this the link you where referring too? https://wiki.ubuntu.com/Touch/ReleaseNotes I dont see anything about screenshots [17:05] hmm, i thought it had something [17:06] here is another link :) [17:06] ikillcypher, http://bit.ly/1d7lly1 [17:06] lol [17:08] ikillcypher, you should add the steps to the wiki page :D [17:08] o.o it is 1am here [17:08] and I have school/work in like 6 hours === gusch is now known as gusch|away [17:08] I will probably continue tomorrow [17:08] but first screenshots xD [17:09] lol [17:09] popey, thanks, we will soon remove share app, and replace phone with the 3 new ones [17:10] can anyone tell me how to take screenshots o.o [17:10] here ikillcypher [17:10] http://vladfilippov.com/blog/2013-02-21-ubuntu-touch-preview [17:10] http://vladfilippov.com/blog/2013-02-21-ubuntu-touch-preview [17:10] lol [17:11] * ogra_ grins [17:11] ikillcypher@ikillcypher:~$ sudo adb devices [17:11] List of devices attached [17:11] that's a problem lol [17:11] mterry, that didn't change much.. we're back to "Ran 0 tests in 0.000s" [17:11] at least it is quick [17:11] lol [17:12] which would indicate the tests may not be installed.. checking again.. this thing has been flash so much ;-) [17:12] ok, yep installed :-) [17:13] ogra_, you have any further ideas for my device? Since that 10 second sleep on the init hasnt been reached. is there a file that get's called like /init thought /init appears to be a binary [17:13] wait my bad wrong file is a script [17:14] yeah [17:14] i would be surprised if it turned into a binary all of a sudden [17:14] can you paste your current touch script ? [17:14] mhall119: there is a "version" field in the manifest file. But the versioning will come as the upgrade mechanism is in place [17:15] ogra_, yeah [17:15] ogra_, let me format /data to see if it works -_- === schwuk is now known as schwuk_away [17:16] annerajb, yes, the initrd was too big to fit into the boot partition, so I had to remove adb. Other than that, hard-coding the data partition device node and enabling the VT and VT_CONSOLE kernel config options was enough (I've heard that's not actually needed though). no time now, back later :) [17:17] Hi i want to make port of ubuntu touch for sony xperia z.Need help === arunprasadr_away is now known as arunprasadr [17:18] ikillcypher, werent you porting to xperia z ? [17:18] lol [17:20] Parth_, i bet ikillcypher would be happy to have someone helping, he just managed his first boot of a buggy port on that device :) === rtg is now known as rtg-afk [17:20] Parth_, he just went away i believe or said he was. [17:20] yes , ill be happy to help [17:23] Parth_, i dont think ikillcypher is around but if you wanna get started looking at his repositories here they are: https://github.com/ikillcypher?tab=repositories [17:23] damn im tired [17:23] Parth_, ported [17:23] check xda soon I will release [17:23] or maybe in afew hours I need to sleep dude [17:23] lol [17:24] ikillcypher: thanks [17:24] ikillcypher, well, at least point him to your code trees :) [17:24] or do your zip [17:24] s/do/to/ [17:24] how do add my name to devices/ubuntu site [17:24] I will do that all later [17:24] chill o.o [17:24] it is late here [17:24] k [17:25] https://wiki.ubuntu.com/Touch/Devices === Namidairo is now known as Namidairo`bnc [17:25] https://wiki.ubuntu.com/Touch/Devices can anyone add ? [17:25] Xperia Z - iKillCypher [17:26] just add yourself [17:26] (you need to log in) [17:27] can you share link to your port [17:28] ikillcypher, you should be able to add to the page there is a edit button at the top [17:29] omg o.o people have pming me for the port [17:29] ubuntu touch gets you famous :) [17:29] ikillcypher, wtf?????????? [17:30] im dozing off here [17:30] o.o [17:30] with what? [17:30] I should thank mamenyaka for error helping [17:30] and ogra_ and sergiusens lol [17:30] heh [17:30] mamenyaka I have been up for more then 24 hours [17:30] and i think Oroku_Saki lol [17:30] and annerajb [17:31] morning school + work + ubuntu-touch porting [17:31] oh come on really now? that's the kind of person you are? [17:31] what? [17:31] I did everything for you, step by step [17:31] and now this [17:32] now what? [17:32] whats wrong with you o.o I dont get you [17:32] there is a special place for people like you in hell [17:32] I dont get this guy [17:32] and now just go to sleep [17:32] what you need? [17:32] a special hell for sleepy people? [17:33] what kind of crap you told these people? [17:33] I really dont get him why is he scolding me for [17:33] I never did o.O [17:33] I thank them for help and you [17:33] Can anyone think of a reason I would be able to get SMS/Calls on ubuntu touch (nexus 4) but not be able to connect to data services? My phone recognizes Rogers as my carrier, but no data. [17:33] ask them [17:34] mamenyaka: be nice :) [17:34] * timp off. [17:34] I'm nice to nice people [17:34] mamenyaka, why did you scold me for o.o [17:34] I thank you for the errors and ubuntu-development team for the help [17:34] errors? [17:35] are you kidding me? [17:35] what else do you want me to do [17:35] kernel errors everything [17:35] he means fixes i guess [17:35] everything is what I did for you [17:35] wtf ?! [17:35] hell, you don't even know how to use cd [17:35] holy smoke o.o [17:35] mamenyaka, ikillcypher credited and praised you here for helping him before [17:35] so you are fighting with me now for credits ? [17:35] wtf [17:36] mamenyaka, yeah as ogra said he mentioned before you where helping him. [17:36] I came here for the past few days trying to set up and got it into brunch [17:36] I just wnat a little respect for spending my whole day with you [17:36] no no, don't get me wrong [17:36] I respected you [17:36] really o.o [17:36] in what way have I disrespected you [17:36] you have made me and thousands of users happy [17:36] really? saying that I helped you error fixing [17:37] yes errors and github help [17:37] really? [17:37] what more do you want me to say? [17:37] I really dont get you [17:37] I said, everything [17:37] you don;t even know how to change directories in linux [17:37] or to use TAB autocopmplete [17:37] ok then thanks for everything [17:37] or adb [17:38] very well [17:38] that's what I wanted to hear [17:38] after all this disrespect [17:38] now go and bath in all your fame [17:38] uh ogra_ care helping out here I think there seems like some misunderstanding going on [17:39] ikillcypher, well, if mamenyaka did most of the work you should say so :) [17:39] well he did help me with alot of stuff [17:39] I respected that [17:39] to me it sounded like he helped you with some kernel fixes [17:39] but I did my part too [17:39] yes, following instructiions [17:40] so both of us are to get the credits here [17:40] yes, you get the credit for having the phone [17:40] but please, it's like a 90% - 10% share of the work [17:40] but I'm fine with that [17:40] lol [17:40] just please don;t tell people that I helped you with some vague errors [17:41] because it was alot more than that [17:41] yes it was mate [17:41] now let break it over [17:41] cause I think thousands of users wont care who got it working but what is working [17:42] they will care who is going to fix all the issues it has left :D [17:43] ikillcypher, just make sure to give enough credit for the work on the wikipage ... and make sure you help with the upcoming bugs that will certainly show up soon [17:43] yes, but there are not so many errors, because our devices share the same repos, except for one [17:43] and I fixed most of the stuff in the common repos on my device [17:44] and he is just using them [17:44] cool [17:44] now I really dont like that [17:44] he is being rude [17:45] just go to sleep please [17:46] come on, can we please just all relax a bit? [17:46] cyphermox, two ofono MRs landing today with you as reviewer... [17:46] don't think we need to do an upload for either [17:46] let's let a couple stack up first [17:47] indeed [17:47] ok! [17:47] thanks! [17:47] having lunch now but ill review right after [17:47] mamenyaka, [17:47] [IMG]http://i39.tinypic.com/2rpv8n6.png[/IMG] [17:47] made some kickass salad :) [17:47] cyphermox, actually though... the first fixes the -dbg packaging, so we might want to upload sometime soon [17:48] ok. i can upload right after review [17:48] cyphermox, sounds good. I have some fresh flounder coming out of the frying pan momentarily [17:48] ;) [17:49] and mamenyaka [17:49] http://forum.xda-developers.com/showthread.php?p=44621291 [17:49] credits given [17:49] tc mate [17:49] cheers everyone [17:49] nights [17:49] awe: sounds good [17:50] ikillcypher, be good, and sorry for overreacting [17:51] im always good my bro [17:51] I have whenever I can [17:51] help* [17:51] ok nights now [17:52] mzanetti, just wanted to give you a heads up that I'll be looking at your PIN/retries bug this week; had a few other bugs stacked up ahead of it [17:52] hello [17:52] hopefully this isn't blocking you [17:52] how do i install Ubuntu touch on an unsupported device? [17:53] hramrach, that doesn't sound like a very good idea [17:53] mamenyaka: manta currently uses mali drivers btw [17:53] if its unsupported, then it requires effort to make it supported. ;)- [17:53] I don't have a supported device so it's the best idea [17:53] not if it doesn't work [17:53] hramrach, which device? [17:54] cubieboard [17:54] hramrach: what's the best idea? You either need to port it or use the community supported instance of it [17:54] !Devices [17:54] You can find the full list of devices, official images, community images, and works in progress at https://wiki.ubuntu.com/Touch/Devices === alan_g is now known as alan_g|EOD [17:54] hramrach, it dosnt appear to be supported by cyanogen mod [17:55] ubuntu touch is based on Android? [17:55] ogra_, why do we flash a ubuntu specific recovery? [17:55] hramrach, it uses the android binary drivers [17:56] binary drivers for what? [17:56] ogra_, i had to remove it because it wasnt working could tha thave to do with my kernel panic? [17:56] sergiusens, hi! what? [17:56] hramrach, devices :P .... [17:56] hramrach, modem, graphics ... etc [17:57] I have native drivers. con't it use those? [17:58] hramrach, what kind of devizce is that ? [17:58] a devboard [18:00] I have a 3D GLES accelerator, a screen and a multitouch tablet [18:00] hramrach, ah ... well, currently the whole setup is built around the assumption that yoou have a minimal android (enough to use the drivers) running in a container and many processes currenntlly talk via libhybris to that container [18:01] so when porting you will hit issues that we didnt plan for yet [18:01] well, you can run andriod on hte board too [18:01] but android is difficult to build becasue it requires certain vintage of ancient compiler which cannt build recent kernel [18:02] well, i guess porting that way would be easier [18:02] swimming with the stream ... [18:02] i think you can surely do a native port but it will likely be a lot more work [18:03] I wonder what kind of android image do I have [18:04] anyway, I also have Adam which is listed as WIP [18:04] so how do I install on somewhat supported device? [18:05] on a fully supported device yoou use plablet-flash from the phablet tools package and it does everything for you [18:05] on a community supported device you grab the android zip and tegh ubuntu rootfs zip and flash them as your device requires [18:06] hmm, anad where is that zip? [18:07] hmm, WIP means there is no image [18:07] the official zips for the four supported arches are at http://cdimage.ubuntu.com/ubuntu-touch/daily-preinstalled/ [18:07] so back to how do I Install on unsupported device [18:07] the unofficial ones are on the devices wikipage [18:08] https://wiki.ubuntu.com/Touch/Install [18:08] there is a manual install section [18:09] note that this assumes a recovery mode that can flash autodeploy.zip files === gusch|away is now known as gusch === jasoncwarner__ is now known as jasoncwarner [18:21] iBelieve, ping [18:21] balloons, pong [18:21] iBelieve, so the file manager had a couple test run snafus, care to weigh some input with me? :-) [18:24] ogra_: community supported devices that are in bold on the wiki can be flash like: phablet-flash community --device [device] [18:24] oh, indeed ! [18:26] balloons, hmm, interesting. are the tests being run on a phone or desktop? [18:26] iBelieve, the tests are run on nexus devices [18:28] jdstrand: just took a stand at opening a click package on manta grep DENIED /var/log/kern.log greps nothing [18:28] ogra_, who's responsible for promoting the image? I thought todays would go through [18:30] pmcgowan, it will, i havent found the time to compare all the test results yet [18:30] pmcgowan, but the next one should be built by now too [18:30] balloons, I'm going to try running that test on the desktop but with the app in phone mode (width = 50gu) [18:31] iBelieve, which test is that? [18:32] test_open_file isn't going to work on the device, so I simply placed a if model() == "Desktop": wrapper around it until support lands. The other two failures though are not so cut and dry [18:32] balloons, any that access places, but specifically the one that failed - ubuntu_filemanager_app.tests.test_filemanager.TestFolderListPage.test_go_home [18:32] sergiusens: so the app failed to launch but there are no apparmor denials? [18:32] iBelieve, kk, ty :-) [18:33] jdstrand: yes, if I remove the aa-exec app confinement from the desktop file it does launch [18:33] let me validate that again [18:33] sergiusens: can you paste the output of 'sudo aa-status'? [18:33] http://paste.ubuntu.com/5986010/ [18:34] jdstrand: com.ubuntu.dropping-letters_dropping-letters_0.1.2.2 is what I want to launch [18:34] sergiusens: which app are you launching? [18:34] or what I'm testing with [18:34] ok, so the profile is loading [18:34] err [18:35] is loaded [18:35] sergiusens: can you perform 'sudo sysctl -w kernel.printk_ratelimit=0' and try again? [18:35] sergiusens: ie, launch it, then do 'grep DEN /var/log/kern.log' [18:36] ah there we go [18:36] jdstrand: http://paste.ubuntu.com/5986018/ [18:36] ogra_: I would like to install by extracting the filesystem to somewhere, not using recovery [18:36] jdstrand: /dev/mali0 [18:36] but I need a working andriod I guess [18:37] which I don't have for either device :/ [18:37] kernel rate limiting ftl [18:37] balloons, it failed for me too, but I figured out what it was. [18:37] hramrach, well, you actually need the android source [18:37] jdstrand: I'll update the bug for documenting purposes [18:37] hmm [18:37] hramrach, grab the tar.gz from the cdimage link i gave you above ... [18:38] so you can at least play with it [18:38] balloons, because list items don't look right on popovers with a Suru them, we switched to using custom Labels, so the Standard list item's text property (which was used to find places), wasn't being used. [18:38] there is no android source for specific device, is there? [18:38] sergiusens: ok, can you add to /var/lib/apparmor/profiles/com.ubuntu.dropping-letters_dropping-letters_0.1.2.2 the following: /dev/mali[0-9] rw, befoer the last brace, then do: sudo apparmor_parser -r /var/lib/apparmor/profiles/com.ubuntu.dropping-letters_dropping-letters_0.1.2.2 and try again? [18:38] sergiusens: this is the complate line: [18:38] iBelieve, so you think you can make a merge to fix both tests? If so, awesome mate [18:38] /dev/mali[0-9] rw, [18:38] hramrach, there are device and non device specific android bits usually [18:39] sergiusens: (ie, we need the trailing comma :) [18:39] hramrach, a working android consists of both [18:39] jdstrand: ack, one sec [18:39] balloons, sure, though maybe not right away. I've already fixed it, I just need the time to upload it and create the merge request. [18:40] iBelieve, k, I'll stand by ;-) [18:40] I see this is going to be a pain-- ideally we would have an out of process helper so the app didn't need direct access to the device (ala pulseaudio for audio) [18:40] 'this' meaning all this device specific accesses [18:41] we can at least get the 4 target devices going though [18:41] jdstrand, like a layer for all device specific access? [18:41] hramrach, the non device specific bits (as well as the device specific bits for the supported devices) are in the git repo from phablet.ubuntu.com [18:41] or rather api. [18:41] annerajb: yes, but it need to be out of process. a linrary layer isn't enough [18:41] s/linrary/library/ [18:41] jdstrand: app opens, I do get other denieds though [18:42] sergiusens: can you paste? [18:42] jdstrand: for /dev/ion [18:42] jdstrand: http://paste.ubuntu.com/5986036/ [18:42] ok, time to search for device specific bits [18:42] thanks [18:43] sergiusens: please add: [18:43] /dev/ion rw, [18:43] jdstrand: just in case the first mali error in there is from the previous run [18:43] sergiusens: then try again [18:43] * jdstrand nods [18:44] jdstrand: success! [18:44] \o/ [18:44] I'll add those to apparmor-easyprof-ubuntu and upload [18:45] great [18:46] sergiusens: thanks for your help :) [18:46] jdstrand: it was all you, I was just a monkey runner ;-) [18:46] sergiusens: ah but a monkey with the required device :) [18:46] lol [18:47] * jdstrand doesn't think sergiusens is monkey at all btw :) [18:47] * sergiusens is not sensitive [18:47] :-) [18:48] ogra_, when i get home i am going to flash the ubuntu recovery because that may have something to do with why my device going into a reboot loop (with ubuntu recovery it just crashed) [18:48] annerajb, ok [18:50] and take a look at the script change see it has no syntax error. apart from that i am really out of ideas -_- apart from commenting kernel_reboot lol [18:51] well, if it fails before the sleep all i can imagine is a typo [18:51] i hope so. [18:51] it worked before ... [18:52] i say that so many times working with code lol. [18:52] it worked (insert time here )ago. [18:52] except that nothing changed in other code apart from the touch script [19:03] awe: your ofono ril patch cleanup branch isn't filed for merge... lp:~awe/ubuntu/saucy/ofono/rilmodem-patch-cleanup [19:04] is that expected? [19:04] I already have the -dbg one approved and applied locally [19:04] cyphermox, yea...my conscience got the better of me, and I decided to do some quick sanity check testing on maguro before proposing [19:04] ;) [19:05] building now, should propose the MR shortly [19:06] guys, unity8.log: http://paste.ubuntu.com/5986119/ [19:07] btw you don't need a .PHONY for the dh overrides [19:07] cyphermox, OK... I got that from sergiusens [19:08] ricmm, or sergiusens ... does any of you have an idea why even though everything on the device runs (SF is up, all processes seem to run fine) the session comes up with a grey/black screen ? mamenyaka cant get his device to work and i'm out of ideas ... [19:09] cyphermox, per https://wiki.debian.org/DebugPackage [19:11] awe: cyphermox I got that from the debian wiki :-) [19:12] sergiusens, you're a bit slow today [19:12] ^^ [19:12] ogra_: hybris calls missing mappings perhaps :-) [19:12] awe: yeah I saw after :-P [19:12] hmm [19:12] ogra_: a quick test is to run test-(the gl ones) [19:12] they work [19:12] mamenyaka, ^^^ right ? [19:13] sergiusens: right. I suspect it was just a force of habit addition ;) [19:13] ogra_: is an upstart job blocking unity from launching? [19:13] sergiusens, i see it in the processlist [19:13] in fact the whole session is up [19:13] powerd/powerbutton seems to work [19:14] ogra_: so, I see something similar on manta for sidestage apps... but I am not focusing time into that [19:14] yeah, understood [19:14] ogra_: grey in full? or does he see something at all coming on the screen [19:14] ogra_: still... I can't think of anything [19:15] ogra_: take a screenshot! [19:15] for real [19:15] ogra_: on the lastest flash if you open the terminal app is it just grey for you? [19:15] sorry [19:15] ricmm, powerd seems to switch backlight on and off ... [19:15] mamenyaka: sounds like permissions problems [19:16] mamenyaka: can you try to run unity8 as root [19:16] okay [19:16] or the glesv2 example [19:16] that runs well [19:16] sergiusens, well, mamenyaka would have to :) i'm not sitting on hs lap [19:16] ricmm: the example works [19:16] ogra_: lol [19:16] what does well mean? [19:16] run unity8 as root [19:16] its failing to find an eglconfig [19:16] that usually means permission issues [19:17] hah [19:17] would you believe that [19:17] it works now [19:17] * ogra_ does belive ricmm [19:17] blindly sometimes :) [19:18] but could I run unity8 not as root? [19:18] mamenyaka: strace unity8 as phablet user then and figure out where the EACCES are [19:18] I mean I have root@ubuntu [19:18] and sort that with correct udev rules [19:18] ogra can shed more light on how to setup udev rules for your device [19:18] yeah [19:19] ogra_: yeah but you also believe in farther christmas elves and unicorns so you're not setting the level high [19:19] (we already set everything with graphics group to 0666, i had hoped that was enough) [19:19] so, should I set everything to 0666? [19:19] mamenyaka: no, you need to strace the failing unity8 [19:19] davmor2, i deny the elves ... but cant deny the unicorns [19:19] and find out what devices are failing to open [19:19] so that you can tune the permissions for them in the rules [19:20] ogra_, what about ogres? [19:20] where do I get strace from? [19:20] apt-get [19:20] and where does apt-get get it from? [19:20] awe, ogres on unicorns, yes, indeed [19:22] mamenyaka: what do you mean? [19:22] # apt-get install strace [19:22] on device [19:22] from the network [19:22] ogra_: no ogres on dragons chasing elves on unicorns ;) [19:22] provision it your network setup with phablet-deploy-networking from phablet-tools [19:23] nope, no elves [19:23] man all this talk of 666 now I can't get the omen music out of my head [19:23] of course from the netwrok, I was suggesting I have no network connection [19:24] no network because of no driver? or because you dont have the UI to set it up [19:24] if its the later, provision the config with the tools [19:24] the wifi was always broken [19:24] otherwise fetch the .deb from launchpad and push it to device [19:25] fetching it is [19:25] davmor2, awe https://lh6.googleusercontent.com/BGmg11iYfZ4iBVhVF-jYTLZSYVjwzTin930rWl1suLQ=w155-h207-p-no [19:26] ogra_, thanks for ruining my afternoon... hopefully will erase that from my brain before dinner [19:26] haha [19:27] * awe prefers rabbit flesh [19:28] is this the one? strace_4.8-1ubuntu2_armhf.deb [19:28] cyphermox, testing looks good, will propose MR [19:28] cyphermox, do you want me to remove the .PHONY? [19:28] or should I leave per the debian wiki? [19:28] or is the debian wiki wrong [19:28] mamenyaka, yeah [19:29] * awe is full of questions this afternoon [19:29] okay, I have strace [19:29] now how do I run it as phablet? [19:30] sudo -u phablet -i [19:31] oh my god, what should I do with all the output? [19:31] awe: cool, thanks [19:32] mamenyaka, strace -o all_the_output.log [19:33] path to spp? [19:33] I did strace unity8 > /home/phablet/log [19:34] mamenyaka, also a good idea would be to read the manpage; strace has all kinds of control for limiting which system calls are output [19:34] yes, even the manpage has a lot of that [19:35] http://paste.ubuntu.com/5986228/ [19:35] mamenyaka, *app [19:36] open("/dev/ump", O_RDWR|O_LARGEFILE) = -1 EACCES (Permission denied) [19:36] there is your prob [19:37] great! [19:38] bu [19:38] ACTION=="add", KERNEL=="ump", OWNER="system", GROUP="graphics", MODE="0666" [19:38] well, check the real file [19:39] what real file? [19:39] open("/dev/ump", O_RDWR|O_LARGEFILE) = -1 EACCES (Permission denied) [19:40] the file it tries to open indeed :) [19:40] ls -l ump [19:40] crw------- 1 root root 243, 0 Aug 14 20:39 ump [19:40] see [19:40] I see [19:41] so do I add an udev rule for /dev/ump? [19:41] so why isnt that 666 ? is your udev rule not processed ? [19:42] it's not /dev/ump [19:42] it's simply ump [19:42] ? [19:42] you mean it doesnt live in /dev ? [19:42] udevadm test [19:43] try that [19:43] and see if it shows any errors [19:43] syspath parameter missing [19:43] thats fine [19:44] i mean the rules it processes [19:44] nothing else [19:44] just some info [19:44] is it a problem if I have more than one ACTION=="add", KERNEL=="ump", OWNER="system", GROUP="graphics", MODE="0666" in the file? [19:45] the last one would apply [19:45] then it's okay [19:45] iirc [19:45] but you should only have one, delete one [19:45] okay, and if there are similar ones, only with different GROUP? [19:45] do the other permissions in dev look like the rule was processed ? [19:46] not really [19:47] http://paste.ubuntu.com/5986270/ [19:47] no, not at all [19:48] how can I check the device name it is expecting? [19:48] how is your udev rules file called [19:48] n8000 [19:48] 70-n8000.rule [19:48] s [19:48] you get it [19:49] getprop ro.product.device [19:49] empty? [19:49] grep ^ro.product.device= /system/build.prop [19:49] no, it's okay [19:49] wrong shell [19:49] ah [19:49] GT-N8000 [19:49] aha [19:49] so is this it? [19:49] yes [19:50] oh well [19:50] I better kill myself [19:50] 70-GT-N8000.rules [19:50] thats what you want [19:50] rebooting [19:59] thank you for letting me waste most of your precious time [19:59] it is now working [19:59] damn, I feel silly [19:59] awesome [19:59] still, font issues, guess I have some other things to do [20:00] isn't there a fix for that? [20:00] I mean apart from reverting the kernel and drivers [20:00] what exactly, you need to be more specific [20:01] well, it's a known problem for some samsung devices [20:01] (fonts to big, upside down, all greek etc) [20:01] font is all messed up [20:01] all greek if you wish [20:01] it's misdisplayed [20:01] ah, yeah, on the galaxy S2 that was solved by going to an older mali driver [20:01] yes, that I want to avoid [20:01] but then, you dont use mali [20:02] because the older driver just needs libandroid and some many other crap [20:02] it is definitely the driver though [20:02] there was I think Ricardo Mendosa researching it [20:02] that we need more memory to allocate [20:02] but never got to the end of it [20:02] could you please ask him? [20:02] well, you talked to him above :) [20:03] oh great [20:03] (he is ricmm ... ) [20:03] ricmm, can you give me all your knowledge about fixing the font issue with samsung? [20:03] haha [20:03] in a can [20:04] thank you really much for helping me all the time [20:04] np [20:04] mamenyaka: whats the font issue with samsung? [20:04] the font misdisplaying [20:04] shows fine on my samsung galaxy nexus [20:04] it's like unreadable [20:04] what device is it [20:04] okay, samsung note 10.1 [20:04] or s2 [20:05] 8000 ? [20:05] yes [20:05] n8000 [20:05] I know about reverting to the odler drivers [20:05] but long ago we emailed about increasing some mem allocation [20:05] mamenyaka, http://people.canonical.com/~ogra/phablet/i9100/img.pngsimilar to that ? [20:06] exactly [20:06] well n8000 uses mali [20:06] I'd say just go with ogra's solution [20:06] ricmm, oh ? it has a pvrsrv device [20:06] I did it once for te unflipped, but it needed libandroid.so and other crap [20:07] if it is indeed mali i guess the rollback is the only solution atm [20:07] for the 10.1 N8000 it is a malii 400 [20:07] trying to find the thread [20:08] mamenyaka: where was this email discussion? [20:08] I am searching for it [20:09] maybe I private mailed you [20:09] after I got the hint [20:10] I found the email [20:11] great! [20:14] mamenyaka: can you try running unity8 with QML_DISABLE_DISTANCEFIELD=1 [20:14] how exactly? [20:15] # QML_DISABLE_DISTANCEFIELD=1 unity8 [20:16] how do I kill of the other unity? [20:16] QXcbConnection: Could not connect to display [20:16] think I got it [20:17] issue remains [20:18] then I dont really know, short of reverting to an old driver [20:18] wait [20:18] if I kill unity8, is it enough? [20:18] edit /usr/share/upstart/sessions/unity8.conf [20:18] what issue remains, the text garbage? [20:18] or unity8 running [20:18] $ stop unity8 [20:18] add: [20:18] will stop the shell [20:19] env QML_DISABLE_DISTANCEFIELD=1 [20:19] right under the respawn [20:19] then just reboot [20:19] stop doesn't work [20:19] stop only works as phablet user [20:19] great [20:20] anyway, just edit the file and see [20:21] how does this vi work? [20:22] okay, got it [20:23] oh now what, stuck at bootlogo [20:24] this is repeating in logcat: I/ServiceManager( 1727): Waiting for service SurfaceFlinger... [20:24] I/ServiceManager( 2087): Waiting for service sensorservice... === rtg-afk is now known as rtg [20:24] heh [20:24] if I reinstall it will work [20:24] you want the fixed powerd that wa uploaded today [20:24] oh [20:25] just add an "and android" to the end of the start on line in /etc/init/powerd.conf [20:25] and it should work on next boot [20:27] hey guys, have some questions as far as porting [20:30] so "start on started dbus and android" [20:30] ? [20:32] yes [20:32] that will make it wait until after the container is up [20:32] still stuck [20:32] okay [20:32] I wait [20:32] currently it tries to access the sensorservice before it is up [20:32] oh, should I have that patch? [20:33] with the sensors [20:33] no, the above change was the patch [20:33] you wait for the android event [20:33] bassicly i fallowed the port system using this link http://forum.xda-developers.com/showthread.php?t=1908008 [20:33] however it still boots into cm 10.1 [20:34] ogra_, 3 min uptime [20:35] still nothing [20:35] but when I fresh install ,everything just works [20:37] well, there is another hack you can do [20:37] reinstall? [20:38] cp /var/lib/lxc/android/rootfs/init.rc /var/lib/lxc/android/overrides/ [20:39] now edit /var/lib/lxc/android/overrides/init.rc ... look for sensorservice [20:39] and change it from "class late_start" to "class main" [20:39] then reboot and see [20:42] this is one messed up device [20:42] nothing [20:42] I mean, same thing, stuck on bootlogo [20:43] even with the sensorservice moved ? [20:43] copied? [20:43] and edited [20:43] yes [20:43] triple checked [20:44] # cat /var/lib/lxc/android/overrides/init.rc [20:44] service sensorservice /system/bin/sensorservice [20:44] class main [20:44] user system [20:44] yeah, thats fine [20:44] then i'm out of ideas [20:44] cat /etc/init/powerd.conf [20:44] author "Michael Frey " [20:44] start on started dbus and android [20:44] well, I will reinstall quickly and edit the QML thing [20:44] yes [20:45] greetings [20:46] i would like to know if anyone has been working Ubuntu on P5110 [20:46] should I downloaded latest saucy or is the current ok? [20:46] aug08 [20:46] okay, wrong question, current updated [20:47] * ogra_ goes for dinner ... [20:47] I've got Ubuntu Touch on a Nexus 4, and I'm thinking about putting the port on my tf101. [20:49] Ubuntu touch its a preview, do we know when the official release will be out? [20:50] awe: did you propose the merge? [20:57] cyphermox, https://code.launchpad.net/~awe/ubuntu/saucy/ofono/rilmodem-patch-cleanup/+merge/180239 [20:57] ivanka, is that you? [20:57] pmcgowan, it is [20:58] but train is arriving so am about to go :-) [20:58] oh too bad [20:58] pmcgowan, you good? [20:58] sure [20:58] ivanka, hope you're enjoying your new gig [20:58] god bless my 300 KB/sec download speed [20:59] pmcgowan, I am, thank you. :-) [20:59] ivanka, well we miss you [20:59] pmcgowan, miss you all, of course [20:59] hehe [20:59] pmcgowan, thank you [21:00] ivanka, things moving along quite well, should have something good for oct [21:01] pmcgowan, excellent! I do keep an eye [21:02] pmcgowan, got to get off the train now - take care and speak soon. All the best! [21:02] ivanka, you too, stop by the channel again! [21:02] pmcgowan, if my computer is on, I am signed in [21:02] pmcgowan, see you round! [21:05] mterry: does seb128 know about account services backing some settings? [21:10] mfisch, he better [21:10] mfisch, i.e. yes [21:23] ogra_, you gotta be kidding me!!!! it works, font issue gone! [21:23] ricmm, font issue fixed! [21:24] ogra_, ricmm noooo, after a little it's getting messed up again [21:28] can you please look at this why wifi is failing to connect? http://paste.ubuntu.com/5986581/ [21:28] [ 132.702546] c1 connect failed event=0 e->status 1 e->reason 0 [21:29] [ 132.702629] c1 CFG80211-INFO2) wl_bss_connect_done : Report connect result - connection failed [21:30] mhall119, is there somewhere I can get a definitive pattern for how to put a flickable in a column and have it and it's contents sized properly? [21:31] I feel like I flail around with anchoring and height and stuff every time [21:36] mamenyaka: what fixed it, my DISTANCEFIELD thing? [21:36] yes [21:36] and then after a while it breaks [21:36] thats weird as hell :) [21:36] but anomalies appear as I use it [21:36] but it suggests memory exhaustion of some sort [21:36] http://kepfeltoltes.hu/130814/screenshot_www.kepfeltoltes.hu_.png [21:37] the word phone [21:37] the keyboard [21:37] awe: done [21:37] thanks dude! [21:38] ricmm, could you please check out the wifi issue I am having? [21:38] sorry, cant really do so right now [21:38] on a deadline [21:39] mamenyaka: but send me an email with the info and I'll take a look at it later [21:39] ricmm, email address? [21:40] rickspencer3: I don't think so, not in the Ubuntu SDK documentation anyway, there might be something for generic QtQuick [21:40] thank you so much [21:40] mhall119, hmmm, there must be a pattern for it somewhere [21:40] I'll find it somewhere ;) [21:41] mamenyaka: ricmm@canonical.com [21:41] rickspencer3: I only recently figured out the right way to use Flickables and Pages together to make the Header disappear [21:41] we could definitely use some code-design guides [21:41] that's some cool stuff [21:42] rickspencer3: if you put your Flickable/Column question into AskUbuntu, we could use it for the Cookbooks on developer.ubuntu.com when it gets a good answer [21:42] (hint, hint) [21:42] :) [21:42] since those are pretty much our code-design documents at the moment [21:43] Hello, is ubuntu-touch a completely different OS or does it run everything you can run on ubuntu itself? [21:45] SrPx: it's the same Ubuntu core [21:46] but not all Ubuntu apps will run on it, because it doesn't use the X11 display server or because they just aren't designed for small screens and touch input [21:46] but all of the Ubuntu Touch apps will be capable of running on Ubuntu Desktop [21:46] so, going forward, new apps should be able to run on both [21:47] and old apps that are ported/upgraded to the new toolkits will too [21:48] rickspencer3: thinking of app for the showdown :P ? [21:48] nik90, schmaybe [21:48] ;) [21:49] I "might" be working on a calorie lookup tool [21:49] mhall119: so I mean, can I run gcc, git, ghc, node.js etc etc on it? [21:50] rickspencer3: hehe [21:50] rickspencer3: if you are, pls add you app at https://wiki.ubuntu.com/Touch/AppShowdownList [21:50] nik90, will do [21:50] rickspencer3: that's where I am tracking the apps being developed for the showdown. [21:50] :) [21:51] about to head out though [21:51] okay [21:51] mhall119: did you know we have a new app called saucybacon :) [21:51] I met a Ubuntu user at the deli during lunch yesterday, going to go get a drink with him in a bit :) [21:51] good times [21:51] rickspencer3: enjoy [21:52] SrPx: I haven't tried most of those, but I believe so. I can (and have) run bzr and gcc on it [21:52] nik90: yeah, the guy who posted it on Reddit was in here, last night I think [21:53] rickspencer3: I love wearing my Ubuntu shirts in public, regularly get approached by strangers about it [21:53] yeah [21:53] my wife hates it though :) [21:53] kinda spoils date nights [21:53] I had my goobuntu shirt, that gets a lot of questions [21:53] mhall119, haha [21:54] I make my wife wear a ubuntu shirt too.. date night saved! [21:54] :-p [21:54] speaking of, I need to report a wishlist bug she pointed out using the phone today [21:59] Will there be a docking station for ubuntu edge? with bluetooth/usb/hdmi/power [22:08] ping boiko [22:08] balloons: pong [22:09] boiko, it seems like calculator no longer lets me use the plus/minus key on the first number [22:09] boiko, I believe it has to do with this merge: https://bugs.launchpad.net/ubuntu-calculator-app/+bug/1210873 [22:09] Launchpad bug 1210873 in Ubuntu Calculator App "Need to enter an operand before using an operator" [Medium,Fix committed] [22:09] balloons: yes, I was discussing this with the developer, but someone else ended up approving the change [22:10] boiko, :-( Well the autopilot tests agree, and started failing this morning [22:10] shall we back it out? [22:10] balloons: either that or open a bug to track the issue [22:11] balloons: btw, I need to get the autopilot tests enabled for the calculator-app CI and autolanding ASAP [22:11] boiko, I'm opening a bug either way. .I have a small fix to the tests I wanted to make [22:11] ogra_, can you send me the diff again i lost the change [22:11] fginther: would it be possible to enable autopilot for ubuntu-calculator-app CI and autolanding? [22:12] boiko, sure I'll give it a test first [22:12] fginther: thanks! :) [22:12] fginther, boiko [22:12] http://91.189.93.70:8080/job/generic-mediumtests/345/ [22:13] beat me to it [22:13] balloons: so, tomorrow I have a meeting with the calc guys, I will talk about this problem (and ask the other guys to be more careful when reviewing MRs) [22:13] balloons: fginther: thanks [22:13] boiko, it seems to be reported: https://bugs.launchpad.net/ubuntu-calculator-app/+bug/1210082 [22:13] Launchpad bug 1210082 in Ubuntu Calculator App "Change of sign operation only works from the second operand onwards" [Low,In progress] [22:14] balloons: but that I think is for that +/- button, isn't it? not for pressing the + or the - buttons, but anyway, this will be fixed tomorrow [22:14] boiko, either way, ok.. thanks :-) [22:15] boiko, have a review if you wouldn't mind tonight.. one line change https://code.launchpad.net/~nskaggs/ubuntu-calculator-app/fix-swipe/+merge/180260 [22:16] balloons: ok, it looks good, I will just wait for CI to run [22:23] boiko, fginther the negative operand bug and the swipe bug fail on the run :-) [22:24] balloons: so, on your MR the mediumtests didn't run, was that expected? [22:24] balloons, was that a good thing? [22:25] boiko, yes, expected because it's not turned on. [22:25] fginther, well a good thing because that's all that failed was the known issue [22:25] balloons: ok, approved then [22:26] boiko, the trouble is we can't enable them now that they are broken.. but had we had them enabled, we wouldn't have pushed in the broken tests and code :-( [22:27] balloons: nah, just enable them, they guys will have to fix the bug before merging any new code ;) [22:28] balloons: or better, wait until tomorrow, I will bring that up in the meeting and after that we enable it [22:35] awe, cyphermox: so I have been finding this week that Ubuntu Touch is utterly unwilling to roam internationally for me automatically, and I can't select a network manually to connect to it. Aside from updating my phone (which I haven't done in a bit, bug #1210347), is there anything else that you'd like me to do testing-wise while I still *am* roaming internationally? :) [22:35] bug 1210347 in system-image (Ubuntu Saucy) "timeout updating from 20130800 to 20130810" [Critical,New] https://launchpad.net/bugs/1210347 [22:36] slangasek, roam mobile-data wise? [22:38] slangasek, 0800 is pretty old in touch terms [22:38] awe: voice [22:38] slangasek, can you bite the bullet and do a fresh install instead of update? [22:38] awe: made a lot of ofono progress, have you? :) Ok, I'll update and see if that resolves anything [22:38] slangasek, my code's being pulled from the archive now finally [22:38] ;) [22:39] and although the -dbg package is currently broken, it will be fixed shortly [22:39] slangasek, once you update, can you please open a bug for this? [22:39] slangasek, we have others canonical folks that are always roaming [22:40] so I know it works [22:40] the ofono scripts are super useful for helping to diagnose such problems as well [22:40] 'specially 'list-modems' [22:40] yes, I can [22:40] you can view the current NetworkRegistration properties which should tell you whether you're registered at all [22:41] slangasek, great [22:42] slangasek, how much longer will you be 'a roaming'? ;) [22:42] awe: until Sunday [22:42] after that, I can try to "roam" with foreign SIMs that won't actually work in the US :) === salem_ is now known as _salem [22:44] haha [22:44] ok. i'm around tomorrow, but off on fri, so if you enter a bug, i'll make sure to take a look tomorrow and maybe we can work on the bug some more [22:45] awe: ok, sounds good :) [22:45] slangasek, have you considered signal strength coverage where you're at? [22:45] just curious... signal str also should be available via list-modems [22:46] anyhow, just grab the output from 'list-modems', and that'll be a good place to start [22:54] awe: it's been a consistent problem when my other phone (android, local SIM) has a connection just fine [22:55] which phone? mako? [22:55] slangasek, hey... I had to ask [22:55] awe: oh, hahaha. I still have ofono disabled with an upstart override since the last bug I was having. :P [22:56] * awe has so many menu choices for what to say next floating across his retinas [22:56] awe: sorry to waste your time ;) [22:57] no worries... you've helped me out many times in the past [22:57] but it definitely made me laugh [22:57] reminds me of one of the first things i ever learned while working on network protocols [22:58] "is it plugged in" [22:58] ;) [22:58] * slangasek grins