[00:04] (Document) https://irc.ubports.com/0SqWtVdJ/file_4016.mp4 [00:48] @Flohack, Congratulations number 432! Your voice was heard and the Marketing Team under Captain Joe (in here) was successfully able to resolve your troubles. Please try clicking the link again and watching the magical URL transformation from wrong..... to right. High fives to @exar_kun === lotuspsychje_ is now known as lotuspsychje [08:23] Good morning people and a wonderful ubuntu day to you! (yes this is an adjective now) [08:37] Good morning Peter, thx [08:45] Leon De Andrade was added by: Leon De Andrade [08:47] Good Morning [08:52] @Leon De Andrade, Hello Leon and welcome! I'm part of the UBports welcoming team. … To help you get started, please take a look right away at our newcomers welcome page (https://ubports.com/page/telegram-welcome) and thanks again for joining us! :) [08:54] Hex Milan, thank you for welcoming :) [08:55] I have often read about Ubuntu touch and I am happy to see that there is still development by the community. Sadly I haven't had the chance to test Ubuntu Touch [09:22] Never is late [09:22] (Sticker, 512x512) https://irc.ubports.com/Rh0QF2u8/file_4019 [11:03] so, who would like to have UBports on Gemini PDA? it should be able to run in theory, now even without arm64 rootfs [11:06] armhf rootfs is capable of running on aarch64 devices, right [11:06] in fact I am running it right now on my secondary phone [11:08] there was an issue with kernel that prevented it [11:17] @vanyasem https://hastebin.com/omuxibicor.diff [11:18] the number of debug prints was pretty desperate [12:21] Anyone from ubports at FOSDEM? [12:23] @technicalbird, I’m not affiliated with ubports, but I am present! Currently at the lightning talks [12:47] The state of Mir live stream FOSDEM https://live.fosdem.org/watch/k4401 [12:48] starts in 2 min [12:51] @alan_griffiths hey you are live ^^ [12:51] wave :D [12:56] awesome thanks ... i cought the flu so fosdem only from the sofa ;( [12:56] Fwd from arudy: The mir talk is going to start :) [12:56] Fwd from arudy: https://fosdem.org/2018/schedule/event/mir/ [12:57] (Photo, 1280x720) https://irc.ubports.com/gB5WTM20/file_4023.jpg [13:10] @TartanSpartan, Yes. its the exact guesture. the second time you tap, dont lift your finger back up. dirctly swipe up and down [13:34] dot [13:34] https://t.me/joinchat/AAAAAAtfczK1cobJMlIXmA [13:35] That's the Ubuntu @ FOSDEM group for Filip [13:38] https://ci.ubports.com/job/xenial-rootfs-armhf/lastSuccessfulBuild/artifact/out/ubports-touch.rootfs-xenial-armhf.tar.gz [13:39] in theory, can this rootfs be used on Ubuntu Tablet or it has different builds? [13:52] @milkor73, and it's never too late also! [13:55] I have read and understood the porting guide but flashing a phone is scary business the guide could use some large warnings about permanent damage and unrecoverable states etc [13:55] @DiogoConstantino, can you find out what technology is powering their live video stream, Diogo? If possible. Unless someone can tell already from their website. thanks [13:59] hey guys why doesn't ubports use terminal app developed by canonical? [14:04] I can ask a friend in the org [14:04] Possibly OBS, plus something else [14:17] @Corne, Any tools that we use to flash aren't dangerous. You won't brick your phone following the guide. [14:17] @Ismaelbonato, But we do [14:19] But it's so different [14:20] the one in Ubports was created by Canonical [14:21] I think you're talking about the reboot version where has actual tabs and has split screen support [14:23] @Flohack Maybe you had answered this question before but... … What do you think about TDLib? … https://core.telegram.org/tdlib [14:24] @SergioSRM, I answered it before and I said, we are aware of this, we are in communication with Telegram guys and we are currently looking into this already ;) [14:24] Because, during the year Telegramw ills hift protocols probably, so all 3rd party clients need to be updated. And this will break a lot for our existing thingie [14:25] Oh! Good and bad news [14:25] (Photo, 720x1280) https://irc.ubports.com/kNJ7mpkL/file_4025.jpg [14:25] @Flohack, Thank you for your answer πŸ‘πŸ˜Š [14:29] I'm using ubports PPA and nothing seems to work :( including clickable [14:30] (Photo, 720x1280) https://irc.ubports.com/pSxcN7UC/file_4027.jpg [14:31] Is that Unity 8? [14:31] Yes [14:31] On my laptop i5 Intel HD mesa [14:32] Libertine and browser either [14:36] I suspect clickable does not export Mir variables, and would need to launch apps with β€”desktop_hint_file=unity8 set [14:37] At the moment the '--desktop' parameter is meant to be used on X11 desktops, I guess [14:37] Would you mind to fill out a report on GitHub? https://github.com/bhdouglass/clickable/issues [14:40] it happen on x11 [14:40] (Photo, 1014x354) https://irc.ubports.com/X7aFxdcC/file_4029.jpg [14:41] @sverzegnassi, Yes I will do this [14:45] @NotKit, Rootfs is same for all devices [14:46] so only Android system.img is different and it contains overrides? [14:46] There's a device tarball different for each device [14:47] where can I check the one for tablet? (probably closed to mine hardware) [14:48] Poke in the json file on system-image server to find it [14:48] frieza? [14:48] Sure [14:51] It has the bits that define grid units for device, and other such things, iirc [14:51] Don't recall if it provides /custom or not, but I think do [14:55] and disables dialer-app/messaging-app for some reason [14:56] Yes, on tablet it hides them, because, well, it's not a phone [14:56] (I supposed all MediaTek SoCs have modem) [14:57] so you can't make calls from it on purpose or it's not supported by hardware? [14:57] Not supported by hardware [14:57] even in some hardware where you have a sim card slot, you still can't place calls [14:58] And we don't have dynamic capabilities detection for specialized apps like that [14:58] @GranPC, Right, baseband has call/sms disabled [14:58] interestingly enough, some basebands only disable calls - sms still works [14:59] (see: nexus 7 2013) [14:59] And I think some don't connect all the pins on hw for it [15:01] I think Telco also disables on their end, for data only sim customers too [15:02] is there any device that uses ofono-rilplugin MTK2 driver? [15:02] or it was prepared for unreleased device? [15:03] But yes, we got complaints about dialer/messaging apps showing on tablet. So I wrote that bit of code to hide them [15:03] No idea on mtk2 [15:05] I suppose it was started for Meizu MX6 Ubuntu Edition/Helio X20 SoC [15:19] @Ismaelbonato, It works now whe I use this "sudo DISPLAY=:0 clickable β€”desktop" [15:29] @Ismaelbonato, I suspect the important factor is that "sudo" is losing some environment variable. A wild guess: is QT_QPA_PLATFORM set? [15:33] rootstock-ng is a pain when you don't have recovery :( [15:40] @alan_griffiths, Sorry I have no idea! I started yesterday to play with this [15:54] [ 5.582677] initrd: checking filesystem for userdata took (including e2fsck) 0 seconds … [ 5.589681] Kernel panic - not syncing: Attempted to kill init! exitcode=0x00000200 [16:10] Hello, I have problem flashing my BQ E5 with Ubports Installer, I'm doing everything like installer says, but it stops on "Please reboot to bootloader" and repeat "debug: Running platform tool fallback exec asar cmd fastboot devices" in console. My E5 is in fastboot mode of course. I also see my device in dmesg and lsusb. [16:11] Beton: i had same issue on the ubports installer [16:11] Beton: i had to install manually from commandline [16:13] lotuspsychje, so I should use old method? Where can I find tutorial, because i don't remember this, and can't find it on Ubports site? [16:14] Beton: sorry, cant remember the proper command either anymore, as i tested so many tools, magic device tool,ubports installer,snap and git.. [16:15] Beton: i can only say, it worked with manually command from terminal and afterwards i had to enable devel channel because of some bugs [16:19] Also, Ubuntu Installer is misleading for BQ E5, because it says "Press Power + Vol Down for reboot to fastboot", but for BQ E5 this is Power + Vol Up... [16:20] Beton: yeah its not perfect yet [16:23] How can I exit from Fastboot mode? [16:25] Power + Vol Up works [16:28] So I must find old documentation now, because I don't want to brick, it's my only phone ;) [16:31] Beton: try to ask here & idle perhaps [16:31] i think UniSuperBox helped me [16:47] Feb 3 16:40:11 ubuntu-phablet watchdog: 'lightdm' (instance '') hit respawn limit - rebooting [16:47] how can I mask lightdm? [16:51] Create the file `/etc/init/lightdm.override` with the content `manual` [16:52] https://bpaste.net/show/c13a191cf9ae [16:52] hm, key 12, I wonder what is it [16:58] Please, is there any group ubpots in portuguese? [16:58] does UBPorts still use forked libhybris? [16:59] Yeah, we've still our own. [16:59] It's better to see what `logcat` says when Mir fails to launch, though [16:59] @UniversalSuperBox, I rebased your tree on upstream already. Its only a few patches different [17:00] You should be able to update it easily [17:00] @UniversalSuperBox, I started with test_hwcomposer and it fails with test_hwcomposer: test_hwcomposer.cpp:238: int main(int, char**): Assertion `eglGetError() == EGL_SUCCESS' failed., which doesn't happen with normal one [17:02] @JBBgameich, Awesome. Can you submit a PR so Jenkins makes a package? [17:06] does it automatically build package for PRs? [17:06] Yes, if there's a Jenkinsfile [17:06] @JhoeyLennon, There it is: https://t.me/joinchat/ALdCegI_knZj301cCAbSMw [17:06] please include https://github.com/libhybris/libhybris/pull/363 in PR then [17:12] @mattbel10, thank you πŸ˜‰ [17:13] Nao tem problema Joey, prazer πŸ˜ƒπŸ‘ [17:22] I found old tutorial and it's working, updated to latest 15.04 RC. I think this should be available on official UBports site, especially of Ubports Installer not working for some older devices... [17:22] https://web.archive.org/web/20170622144804/https://wiki.ubports.com/wiki/How-to-flash-existing-ubuntu-touch-devices-with-Ubports-images [17:30] I see you have an irc bridge might someone care to share the domain and room I much prefer IRC over telegram [17:32] Freenode #ubports [17:33] The Community Update is timed to start in two hours, is that correct? [17:38] yerp [18:34] @NotKit, Waait. There is a process that can reboot the phone if another process respawns too often? [18:48] maybe it's just upstart? [18:57] @Flohack, yes [18:58] Looking to join the Community Update Live? It's starting right about.... NOW. Get the link at the News Channel and join us and learn more about your UBports community. https://t.me/ubports_news [18:58] @dohbee, Could tjat explain the unforeseen reboots of FP2? [18:59] maybe? [18:59] i mean, it would be in the logs i guess [19:00] I think I could do a fix on one of my community update questions in the forum, what is the common hardware aspect linking cooler freiza and turbo which allow Anbox on them but currently not so on other devices? [19:00] @NotKit, no, there's a watchdog script that reboots phone if things keep crashing [19:00] @Flohack, No, that's kernel problems [19:01] so to build libhybris by hand, better make a copy of UBports rootfs, install build packages and compile it? [19:01] @NotKit, you want to build it on the device? [19:02] I guess that's the fastest way? [19:02] maybe [19:02] @NotKit, https://askubuntu.com/a/623311/50737 [19:02] WILL! I've got yhour back! [19:02] release your inner Durst [19:02] ie "make a chroot" [19:03] WE ARE LIVE ON COMMUNITY UPDATE SO EXPECT THIS CHAT TO SEEM WEIRD IF YOU ARE NOT WATCHING [19:03] @wayneoutthere, as opposed to normal weird? [19:03] @TartanSpartan, Are not qualcomm snapdragon device, in fact are mediatek. There is an issue with qualcomm chipset. … Not that I want to answer instead of a F4, they can do too, but your curiosity is satisfed now. [19:04] @wayneoutthere, \o\ /o/ \o\ /o/ [19:04] Ok so I think I phrased my question correctly in that case. [19:04] guess i'll pause the Shalamar for a while [19:04] so MediaTek supports it, cool [19:05] @TronFortyTwo, turbo is not mtk, it's exynos [19:06] there's nothing specific about the hardware i guess. turbo has the most RAM of supported devices i guess, and cooler/frieza are exactly the same except for screen resolution [19:06] do that again, Jan [19:07] WILLLL! [19:07] get him! [19:07] haha\ [19:08] @dohbee, Oh you're right, [19:08] Was that a donke sound or a city name? [19:11] "The Graveyard of my Failed Dreams " - Will Atwood, 2018 [19:12] and there are 2 canadians now too [19:12] haha [19:13] ;D [19:14] UBports Community Steering Committee "Doing business as - "UBAM" [19:14] @Flohack airports are the best place to join from :P [19:14] love you man. [19:14] β™₯️❀️ [19:15] UBAM = UB AMazing [19:16] UBFR = UBports First Responders [19:17] HIGH FIVE FAQ WILL [19:18] 😻😻😻 [19:18] Very nice to see Will in the fold there, he's a stand up guy. [19:18] Yeah, very good words [19:18] we have our new stage man [19:19] well said. this will be available as audio to grab/go. [19:20] panopticon jokes probably not the best thing to make, concerning production of a phone os [19:22] ubuntu touch is not... [19:23] not a word made sense. [19:25] so all the renaming will have to be renamed back to unity8? [19:26] Call to action: Anyone with C+, C++, or with Unity, come! Help! :) [19:27] Want convergence sooner? Come! Help! :) [19:30] lost you marius :) [19:32] Jan - be his spokesman [19:32] does UBPorts Anbox use Android graphics drivers directly, like Alien Dalvik/SFDroid, or through transport layer? [19:34] I'll ask again guys, what's the timescale you're thinking for Anbox and further rollout? [19:39] Thanks Dalton, good to know :) [19:40] (Photo, 2560x1440) https://irc.ubports.com/KcYuYe7H/file_4035.jpg [19:40] Ubuntu and Ubports users at FOSDEM salute you all! [19:40] \o/ [19:40] hey guys at FOSDEM. [19:41] Whats with the sunscreen bottle? [19:41] It finds its way to all the photos :) [19:43] We nees to be prepared for Sunny Brussels! Right @arudy ? [19:45] @mariogrip you having a solid version of Xenial with Anbox as q daily driver is just envy inspiring :p [19:46] *a [19:47] Is Nexus5 16.04 at same stage? [19:48] you should be able to generate automated "changelogs" [19:48] Btw Dalton if you're unsure about the pronunciation of my name, "shayyy-mus" :) [19:48] muhs or moos [19:48] wait [19:49] facepalm [19:49] Former. I'll give you a fistbump if you say it live just now :) [19:50] Sweet info. [19:50] well changelogs are never user friendly [19:50] go read the updates for apps on android [19:51] you can just do `echo "Bug fixes and performance improvements." >> changelog-$(date +%s).txt` [19:52] write better commit messages :) [19:53] Boom! Love you guys. [19:54] one. two. threeve. [19:54] Kwak [19:54] quackly. answer [19:55] It was not quick. It was quack. [19:55] Any timescale for general functionality of Libertine on Xenial? [19:55] dot biz [19:56] dot im [19:56] dot phone [19:56] did you say 'slap a bun on it'? [20:00] *general [20:00] it works, sort of [20:00] @Crash_Burn you can install it in a click if you want i guess [20:01] trust no apps. [20:01] Libertine works on Vivid for me, not so on Xenial. [20:03] Yes you can sir! #FellowAmerican [20:04] not really [20:04] the browser should be confined [20:04] I was saying yes he can pronounce my name [20:04] webapp-container is for sure. i think we confined the browser too [20:05] but it is a deb [20:05] Then I will poke Marius in PM to make him poke the app hehe. [20:06] "Yumi is my bae" [20:07] Thanks for all the answering brothers! [20:07] Rebranded! Great job crew [20:08] hahaha [20:08] does anyone know the facebook url?? haha [20:08] no. [20:09] πŸ‘ [20:09] Whoops, I hit "End Call" rather than "Go offline" XD [20:09] I tried to go but I got drawn back [20:09] so yeah, browser downloads are a bit convoluted [20:09] Thanks guys again!! πŸ‘ŠπŸ˜Ž [20:09] because it's a phone where the filesystem is meant to be abstracted away [20:10] @UniversalSuperBox My daughters wonder what that snowman is all about. It's very important feature in your backdrop now. Maybe you can dress it differently each week? :) [20:11] ...but if the internal storage's /Downloads folder works for some people, shouldn't we permit /Media/SDcard/Downloads to work for others? That's one of the power-user features which attracts people to Firefox. [20:13] @wayneoutthere, It's kinda a safety for the things in that basket [20:14] just in case [20:14] @DiogoConstantino, Oh yeahh 🀘🌞 [20:15] @TartanSpartan, Did I say we shouldn't? [20:15] No, but just to reaffirm my point. [20:16] then please don't reply in a manner which implies i did :) [20:17] i would say firefox is probably less secure in this manner. but really i'm just pointing out where the difficulties lie [20:18] but the webbrowser-app is less secure in other areas (it's based on outdated chromium and lacking many security fixes) [20:18] Just in case anyone had ambiguity there. Sorry if I came across as contrarian. [20:21] Is stats.ubports.com down, or moved to other domain? [20:21] firefox and chromium kind of make me wish i never had to use the web again [20:22] @Beton, it was taken down intentionally [20:24] Why? I think its good idea to have statistics like this [20:27] It was extremely inaccurate and people were quoting it [20:29] @UniversalSuperBox, So, you know the real numbers... [20:29] Nope [20:29] Because the stats were wrong [20:30] I forgot to ask just now: any development on the unav situation Florian mentioned the previous time [20:30] The aggregate were correct [20:30] any news there? [20:30] https://twitter.com/costalesdev/status/958045845529792517 [20:30] I think Marius is right that if the inaccuracies are the same in relation to each different device, then we know that the ratios between the devices are accurate [20:31] @UniversalSuperBox, What it means? [20:31] The percentages per country, etc. were more correct. [20:32] ok [20:32] If they are all out by a factor of x5, the proportions are unchanged [20:33] Because method of colecting statistics was wrong, it based on asking for updates iirc [20:34] it should base on uniqie device ID, every Ubuntu Touch device have it [20:34] Crash reports base on this too [20:35] Also I see that there still is option to send crash raports to Canonical on latest 15.04 RC, shouldn't it be removed? [20:37] @Beton, i don't think so. crash reports are still useful, especially once 16.04 is made stable [20:38] well, it was naive of me to expect Mir to just work: https://bpaste.net/show/15afcdb69c71 [20:39] I updated libhybris though and test_hwcomposer runs now [20:42] key 12 would be NATIVE_WINDOW_DEFAULT_DATASPACE https://github.com/libhybris/libhybris/commit/ad5b3d2530d6bd534930c43ee9f34666bc7b1808#diff-e5a142e46157d2a0ad4735bb1346de52R304 [20:43] @dohbee so Canonical still support Ubuntu Touch or these reports go to UBports developers now? [20:43] should I build xenial_-_mir28 branch of mir-android-platform? [20:44] @Beton, Canonical doesn't exactly, but the phone image is still Ubuntu, and so crash reports should go to Ubuntu. [20:44] i wouldn't suggest that ubports should run their own crash report server at all. [20:45] ok, i understand now [20:48] btw, I flashed my BQ E5 today, and I saw that I can choose 16.04/dev branch, so something changed about 16.04 version on legacy devices? [20:51] it's been available for a while now [20:54] @dohbee, Legacy devices only. [20:55] @Cesar_Herrera, Huh? [20:56] Only those were from Canonical. [20:58] Canonical does not support any of the phone/tablet devices any more [20:59] Yes. [20:59] no, the m10 too [21:00] but really, whatever you are saying has nothing to do with what i was saying [21:00] I shoud say Ubuntu and no Canonical. [21:01] i still have no idea what you're trying to say [21:01] Ubuntu is still Ubuntu [21:02] many of the underlying packages still come from the main ubuntu archive. ubports doesn't rebuild 100% of the stack independently [21:02] I refered about the crash reports. [21:03] ie systemd/upstart comes from ubuntu. glib comes from ubuntu. pam comes from ubuntu. [21:03] yes, and if on 16.04 systemd crashes, it would be nice if ubuntu would fix it :) [21:04] :) [21:05] The following packages have unmet dependencies: … libmirrenderer-dev : Depends: libmircommon-dev (= 0.26.3+16.04.20170605-0ubuntu1.1) but 0.24.20ubuntu11+ubports+0~20180125235309.11~1.gbp713a61 is to be installed [21:05] where can I get the correct version of libmirrenderer-dev? [21:06] i think plasma mobile crashes go there too [21:07] oh bummer, i can't actually see the reports any more it seems :( [21:14] @UniversalSuperBox, Who????? I'll punish the man! ;) [21:15] wait, so mir-android-platform repo is for newer mir versions (basically it got split for main Mir tree), and UBPorts xenial still uses 0.24? [21:16] yep [21:17] I get somewhat confused by all the branches, any quick guide on that? [21:17] what repo should use for device tree based on cm-12.1, ubport-5.1 or halium-5.1 ? [21:25] alecsito was added by: alecsito [21:42] @alecsito, Hello Alex. Welcome! Check https://ubports.com/page/telegram-welcome … Not-a-botβ„’ Ask, chat, explore! [21:44] @Walid, halium [21:45] Just know that it won't be officially done until... probably after OTA-4 [21:47] You can join me on my quest to flesh out the Halium 'standard' though, along with the guys from Plasma Mobile and LuneOS [21:50] what was the newest device to get relatively functional port though? [21:52] Walid ported the Moto G3 using the ubp-5.1 base [21:52] I got middling results on the Moto G5 Plus using halium-7.1 [21:52] But now I'm working on the 5X and hitting a brick wall [21:53] on initial startup level or somewhere later? [21:53] Later. Console programs work for a bit, then something gets printed in the log and everything stops launching [21:53] Let me find that... [21:54] `[ 25.613724] msm_rng:Data not available!` [21:54] After that, I can only launch a few programs. New shell sessions won't appear. [21:55] I had weird freezes with Halium on Qualcomm devices just like you described, but it happened later [21:56] @NotKit, I got a somewhat working port on Samsung S3 Neo πŸ˜† [21:56] But only on 16.04, 15.04 has lots of issues with that [21:56] Again, only happens after a certain amount of things have been done. For example, after I tossed in Telnet rather than ssh on boot, it'll run slightly longer. [21:56] Until I try to read too much from `/dev/random` [21:57] I don't know if that's a cause or a symptom though [21:57] Can't use sudo [21:58] So most things go out the window. :P [21:58] does it happen only on Ubuntu rootfs or on Halium reference as well with your kernel configuration? [21:58] I haven't switched to reference yet. Going to try that now. [22:08] Just waiting for that sweet systemimage build... [22:19] @UniversalSuperBox, Both? [22:20] @dohbee, that would be disastrous [22:21] Well, loss of entropy means the rng gets consumed, and then you get the freeze [22:21] Oh! You were serious [22:21] Who knows why the rng kills itself, though... [22:22] Needs some special entropy generator from Android? [22:22] It's in-kernel [22:24] Hopefully. If it was in an Android service we'd have some pain coming up [22:24] @vanyasem did you get similar behavior on your build? [22:33] @dohbee, And then it will consume the battery. Finally found the reason for aging batteries [22:34] I don't think that's related [22:35] Hahah joking [22:40] and then it leacks memory and here the famous RAM problem source found too [22:40] (😁) [22:51] About run unity8 on top of x11! When will it be ready? [22:55] Mir already works on top of X [22:55] I never tried thought [23:00] @dohbee, can it be disabled though? [23:01] @UniversalSuperBox, exactly the same [23:01] @NotKit, Sure I guess [23:06] @NotKit, yep, same problem with reference [23:09] oops [23:09] (Photo, 1280x960) https://irc.ubports.com/KH0y61T4/file_4037.jpg [23:10] 16.04 huh [23:10] https://bobsummerwill.com/2018/02/03/helping-purism-not-to-make-the-same-mistakes/ [23:11] so it implementing NATIVE_WINDOW_CONSUMER_USAGE_BITS was enough to get Mir running on Helio X27 [23:11] I wish rebuilding Mir with few lines added didn't take few hours though, lol [23:12] @UniversalSuperBox, but Nexus 5X is like a reference for Halium? so it can be due to kernel config differences? [23:12] @NotKit, mir is a beast, but if you want you can send a pr that will trigger a build on our build system 5-6minutes to build [23:12] will do [23:13] our 64-core arm64 flexes :) [23:13] @NotKit, It did it with and without the changes [23:13] wow, way better than poor OnePlus 2 I used for building, lol [23:13] @mariogrip, hey Marius :P. I just wanted to inform you that the latest ubports halium rootfs doesn't seem to work. it just bootloops with no graphics [23:14] @vanyasem, humm, noted, will take a look at that [23:14] @mariogrip, I assume that's due to caf-related changes [23:14] Since we're mobbing @mariogrip... ever had a deadlock due to the msm_rng failing? [23:14] I tested on HTC 10 btw [23:15] it's latest rootfs on photo though [23:15] @UniversalSuperBox, nope [23:15] @NotKit, is that a caf device? [23:15] but I built hybris from @JBBgameich tree [23:15] no, MediaTek [23:16] yeah, it's only qcom caf's that has some struggles [23:16] had to revert some stuff due to our samsung chip device (pro5) didn't like the new hwc changes [23:16] but im working on a workaround for that [23:16] MTK wouldn't like them too with prebuilt HWC, I think [23:17] mediatek is smart, they don't change enums values in the middle of the DAMN enum [23:17] Ah, so we're looking at separate rootfs's again? [23:17] but if size of complete struct changes by additions in the end, wouldn't it still break? [23:19] @UniversalSuperBox, nope, just need to dome some magic with mir to have two androd platform binaries, one 7.1++ and one for 6.1β€” and then load the correct one on init [23:19] like pulseaudo does [23:19] @mariogrip, getting mtk_disp_mgr 32-bit ioctls compat layer running was crazy enough to question that [23:20] since we cannot rebuild the android stuff on the pro5 we cannot use my other fix that halium can use :( [23:20] (Photo, 1280x960) https://irc.ubports.com/0grjeZZH/file_4039.jpg [23:20] Guys my 5X needs help [23:20] since that reqires changes to hwc on android [23:20] @UniversalSuperBox, I think the 808 turned into a fusion reactor, gg [23:20] you just solved free energy [23:20] @Lyokanthrope, Can I patent this [23:21] @mariogrip, yeah, i assumed that as well as i only got issues on a CAF HTC 10. older version of this rootfs works perfectly [23:21] anyway, to get touchscren rotated, will LIBINPUT udev rules apply? [23:23] The plot thickens. My 5X works with my kernel config changes in reference. [23:23] (It was booting the wrong rootfs before) [23:23] @NotKit, Rotated? Did you go through the initial setup? [23:24] I didn't yet since touchscreen is misaligned [23:24] @UniversalSuperBox, same expersience. reference rootfs worked, but not ubports one [23:25] @NotKit, Pretty sure initial setup is always portrait, and rotation will work normally after [23:25] @dohbee, the screen is physically rotated iirc. so the image needs to be rotated 180 degrees all the time [23:26] Huh? [23:26] that's a phone with a physically flipped screen iirc [23:26] @vanyasem, the kernel *should* do that, should not be a userland issue [23:26] (at least qcom does that) [23:26] @mariogrip, @NotKit ? [23:27] I'm pretty sure it's a tablet? [23:27] @vanyasem, yes [23:27] @dohbee, (the screen itself is rotated by factory) [23:27] well, actually, the screen is portrait [23:28] I still don't know what that means [23:28] but touchscreen is reported landscape [23:28] @dohbee, well that doesn't really matter. it's a portriat tablet, counts as a "big phone" [23:28] Does not compute [23:28] like the oneplus 5 where the screen is monted upside down [23:28] and turned around in software (kernel) [23:29] Still seems irrelevant [23:29] the device is "PDA" though, so neither phone nor tablet technically [23:31] The setup wizard is always in portrait. So if it's appearing portrait, it doesn't mean rotation isn't working [23:31] no, it's not the issue [23:31] it's that touch screen is reported landscape [23:31] Ah [23:32] So a bug, or maybe something not synced [23:32] it's "a feature" [23:32] since basically for Android they set ro.sf.hw_rotation=270 [23:33] Oh, change that maybe? [23:33] well, it needs to be undone somehow, but my original questions was, would Mir handle libinput calibration set through udev rules? [23:33] or this need to be fixed in kernel? [23:34] I think it should work [23:45] yes, 270 degrees rotation matrix helped [23:47] well, UI runs, everything else doesn't :) feels way smoother than GNOME Shell on same device though [23:50] Nice [23:52] does anyone know how /dev/wmtWifi is controlled on frieza/other MediaTek devices? [23:56] What level of control? NetworkManager is used on userspace side