| T42 | ※Téléаgrаm ♀♀GrouР members Рrivаte Рromotio♀♀ %lastname% was added by: ※Téléаgrаm ♀♀GrouР members Рrivаte Рromotio♀♀ %lastname% | 04:53 |
|---|---|---|
| T42 | <edp_17> @elros34 : I have built and installed your qt5-qpa-hwcomposer-plugin (external_display) to my Nexus 5. The device starts and works normal (so far) but when I plug in the htmi cable, nothing happens. Is there something else I need to add? (Into droid-hal-device.conf maybe?) Thanks. | 08:02 |
| T42 | <Andy %lastname%> @wdehoog [I promise I'll do so after releasing from test …], @wdehoog Personally want to say thanks. Got a nice email this morning saying the Bacon 3.4.0.24 has been built already - Brilliant news - thanks for the hard work and sticking with it. Flashing it later today!😁 Happy to help reporting bugs if that helps | 08:34 |
| T42 | <Andy %lastname%> Sorry should read not flashing but upgrading OTA! | 08:34 |
| wdehoog | Andy: have fun | 08:45 |
| wdehoog | @mal how can I help you debugging libhybris for cm12.1 ? | 09:04 |
| piggz | mal: looking at the amount of PRs in hybris-boot, maybe someone could look at https://github.com/mer-hybris/hybris-boot/pull/189 ? | 09:15 |
| mal | wdehoog: does gdb show anything useful? | 09:25 |
| T42 | <Andy %lastname%> (Document) https://irc.thaodan.de/.imgstore/qIszSG0i1k.png | 09:28 |
| T42 | <Andy %lastname%> (Document) https://irc.thaodan.de/.imgstore/hDAyjjF07w.png | 09:28 |
| T42 | <Andy %lastname%> (Document) https://irc.thaodan.de/.imgstore/BZNGHIu9Zn.png | 09:29 |
| T42 | <Andy %lastname%> (Document) https://irc.thaodan.de/.imgstore/RNBkVFZzaG.png | 09:29 |
| T42 | <Birubiruungu> @Andy %lastname% [<reply to media>], 👍 | 09:31 |
| T42 | <Andy %lastname%> Well enough only to need one screen shot! Sorry | 09:36 |
| T42 | <Andy %lastname%> Next trick is to spend enough time now to nail @wdehoog boot screen injector - never worked so far so time saved updating not reinstalling to 3.4.0.24 will be spent building an injector! | 09:47 |
| T42 | <Andy %lastname%> https://github.com/wdehoog/sailfish-bacon-images/tree/master/boot-logo | 09:47 |
| wdehoog | dear Andy I have no idea what you mean | 09:49 |
| T42 | <Andy %lastname%> I am trying to build the boot logo script on the page https://github.com/wdehoog/sailfish-bacon-images/tree/master/boot-logo | 09:50 |
| wdehoog | @mal please bear with me I have NO idea what to do. How to build libhybris with debug info, how to install it or how to use gdb to get info | 09:50 |
| T42 | <Andy %lastname%> So i will then have a decent logo to match the phone (currently have an old Ubuntu touch logo) | 09:50 |
| T42 | <Andy %lastname%> It just needs time | 09:50 |
| mal | wdehoog: OBS automatically generates the debuginfo packages, locally you can do mb2 ... build -d | 09:54 |
| wdehoog | @mal so if I build it on OBS I can just install it on an existing installation (no need to build and install a complete image)? | 10:05 |
| T42 | Amy Christy Sweeney %lastname% was added by: Amy Christy Sweeney %lastname% | 10:08 |
| mal | wdehoog: yes, it should be ok | 10:25 |
| mal | I have tested libhybris many times with only that rebuilt | 10:26 |
| wdehoog | @mal installation of 0.0.5.38 done, rebooted, sort of works, feels sluggish, journal: http://sprunge.us/XdrnMg | 10:59 |
| wdehoog | logcat: http://sprunge.us/enM1tv and dmesg: http://sprunge.us/20OnuL | 11:03 |
| mal | a lot of egl issues in logcat | 11:27 |
| T42 | <elros34> @edp_17 no, it should just works. Your external display should show that something is connected with blank content and when something change in your primary display then it should start working. Check journal, you should at least have detected hotplug event and logcat. What was the version of hwcomposer on that device? | 11:53 |
| T42 | <elros34> ah and also it didn't wor for me on latest libhybris so maybe that is the issue. | 11:54 |
| T42 | <elros34> @wdehoog, did you update also qt5-qpa-hwcomposer-plugin rebuild against your updated libhybris? IIRC that was important when testing libhybris failure. | 12:00 |
| wdehoog | elros34: I have build latest libhybris on OBS and installed from there, I think it rebuilds and install qt5-qpa-hwcomposer-plugin as well | 12:10 |
| T42 | <edp_17> @elros34: I don't know the version of the old qt5-qpa-hwcomposer-plugin as that was built for 3.3.0.16 a while ago. Since I upgraded and re-built my local repo to 3.4 but not yet installed onto my device. So the version was whichever version was generated for 3.3.0.16. How can I check? | 12:16 |
| T42 | <edp_17> I'll see the journal to see the events then come back. Thanks. | 12:16 |
| wdehoog | I deleted .jolla-startupwizard-done and .jolla-startupwizard-usersession-done and the wizard successfully ended | 12:24 |
| wdehoog | rebuild image using latest libhybris. wizard successfully ended | 12:52 |
| wdehoog | vibration does not seem to work (I notice nothing) but devel-su test_vibrator does work | 12:57 |
| mal | wdehoog: did you update system/core before you rebuilt android side? | 13:08 |
| mal | https://github.com/mer-hybris/android_system_core/commits/hybris-12.1 | 13:08 |
| mal | if you even rebuilt it | 13:09 |
| T42 | <elros34> @edp_17 I was asking about hwcomposer version reported in journal not qpa plugin. Also I use older libhybris: https://github.com/elros34/libhybris | 13:17 |
| T42 | <ankaos> fixed : # Requires: droid-config-oxygen-pulseaudio-settings | 13:25 |
| mal | @ankaos why do you need that? | 13:40 |
| mal | I mean why is it commented out? | 13:40 |
| T42 | <ankaos> mal: fixed Warning[11/25 18:23:53] : repo problem: nothing provides pulseaudio >= 12.2+git8 needed by pulseaudio-modules-droid-hidl-1.3.0+master.20200525095923.a35a5ba-1.armv7hl | 13:53 |
| mal | are you sure you use correct target? | 13:53 |
| mal | what does "sb2 -t $VENDOR-$DEVICE-armv7hl -m sdk-install -R ssu re" say? | 13:54 |
| T42 | <ankaos> mal: Device release is currently: 3.4.0.24 | 13:59 |
| T42 | <edp_17> @elros34 : I have checked journal for version of hwcomposer but haven't found there. These are the related lines in the journal: https://paste.ubuntu.com/p/c4bFVcbF6y/ | 14:01 |
| T42 | <edp_17> Oh, sorry: Version: 1030001 | 14:02 |
| mal | @ankaos how do you build the image, did you specify correct version when building it | 14:02 |
| T42 | <ankaos> mal: yes | 14:02 |
| mal | the RELEASE export | 14:02 |
| mal | 3.4.0 has that 12.2+git8 | 14:02 |
| mal | check sb2 -t $VENDOR-$DEVICE-armv7hl -m sdk-install -R zypper if pulseaudio | 14:03 |
| wdehoog | @mal indeed I did not. which step in the build process or command does that? | 14:06 |
| mal | check sb2 -t $VENDOR-$DEVICE-armv7hl -m sdk-install -R zypper ref | 14:08 |
| mal | I mean first that | 14:09 |
| mal | then sb2 -t $VENDOR-$DEVICE-armv7hl -m sdk-install -R zypper dup | 14:09 |
| wdehoog | @mail I meant system/core. is that git bladibla and the build_package.sh -d ? | 14:10 |
| mal | wdehoog: first "make hybris-hal" as usual | 14:12 |
| mal | wdehoog: I mean after updating from git first | 14:12 |
| T42 | <ankaos> mal: https://paste.ubuntu.com/p/B4N4mmpYfM/ | 14:16 |
| mal | wdehoog: maybe update hybris-boot repo also while at it, in case you haven't done that | 14:16 |
| wdehoog | @mal my hybris-boot ends on dec 2016 and is 'Not currently on any branch' should I update to HEAD:master ? | 14:24 |
| mal | yeah | 14:27 |
| T42 | <ankaos> hı. open telnet. ı read journalctl output: https://paste.ubuntu.com/p/C33MT79PFT/ . device not open screen | 14:27 |
| mal | @ankaos which android base? | 14:28 |
| T42 | <ankaos> Lineageos 14 q | 14:28 |
| mal | is /system partition mounted? | 14:30 |
| mal | these suggest no "Could not import file '/usr/libexec/droid-hybris/system/etc/init/atrace.rc'" | 14:30 |
| mal | so probably it failed to create mount services during droid-hal build | 14:31 |
| T42 | <ankaos> mal: system mounted. | 14:32 |
| T42 | <ankaos> mal: ı have atrace.rc | 14:33 |
| mal | are those symlinks ok on device? | 14:34 |
| mal | check ls -l /usr/libexec/droid-hybris/system/etc/init/atrace.rc | 14:35 |
| T42 | <ankaos> mal: sh-3.2# ls -l /usr/libexec/droid-hybris/system/etc/init/atrace.rc | 14:35 |
| T42 | <ankaos> lrwxrwxrwx 1 root root 26 Nov 26 2020 /usr/libexec/droid-hybris/system/etc/init/atrace.rc -> /system/etc/init/atrace.rc | 14:35 |
| mal | @ankaos so you /system/etc/init/atrace.rc ? | 14:35 |
| mal | +have | 14:35 |
| T42 | <ankaos> mal: compositor error? | 14:36 |
| T42 | <ankaos> mal: https://paste.ubuntu.com/p/zXP3fMth3s/ | 14:36 |
| mal | those droid-hal-init import errors are important part now | 14:37 |
| mal | Apr 21 21:22:55 mi_max2 systemd[1]: system.mount: Cannot add dependency job, ignoring: Unit system.mount failed to load: Bad message. See system logs and 'systemctl stat | 14:38 |
| mal | you said you have system mounted but that says system.mount has some issue | 14:38 |
| T42 | <ankaos> ı install los-14.1 | 14:39 |
| mal | ? | 14:41 |
| T42 | <ankaos> reinstall sfos | 14:43 |
| mal | no idea why you would reinstall lineage os also | 14:49 |
| T42 | <ankaos> :( | 14:56 |
| mal | you might be misunderstanding how things are related on the system | 14:57 |
| wdehoog | @mal csd vibrator test now works. thanks. | 15:21 |
| szopin | I'm in: Welcome to the Mer/SailfishOS Boat loader debug init system. | 15:31 |
| szopin | sadly only getting logcat read failure | 15:36 |
| T42 | <elros34> you must be faster, do it as soon as droid-hal-init start and before it fail | 15:48 |
| T42 | <elros34> or maybe it didn't fail, I am not sure anymore. Check in journal | 15:49 |
| szopin | would init.log, dmesg output and journalctl be helpful too, before I try to get the logcat faster | 15:51 |
| szopin | can pull those at least | 15:51 |
| T42 | <elros34> init.log is not useful anymore | 15:51 |
| szopin | got it: https://privatebin.net/?d1d86aabf36a4c44#38QEyght3EXrav8j9ddwwcuaCTi8rU4Tj7RqEaxquowi | 15:56 |
| szopin | journal https://privatebin.net/?94428551a32858d3#BwiMSd51CAGumfzembH5a2jfE25BcPDRzG3VNMN4qp9y | 15:57 |
| szopin | and dmesg https://privatebin.net/?10f1e8f64b6f4d19#7vZtrVsJuFw79XYVCPaBhAws8aaMgpfzdMKc5DwV3Dxv | 15:58 |
| T42 | <Eric W. Thomas %lastname%> You save my life out of debt and I don't think I could ever repay you for all that you have done for me, I will keep telling people about your good work in my life. | 16:02 |
| T42 | <Eric W. Thomas %lastname%> https://t.me/joinchat/AAAAAEPBAcc3KJcoln1pbw | 16:02 |
| T42 | <elros34> szopin: first line in logcat is something to fix | 16:03 |
| szopin | hmm does it need recompiling with some extra parameters or... | 16:07 |
| szopin | ok I think I found what needs to be done | 16:20 |
| mal | szopin: is that which android base? which device? | 16:31 |
| szopin | cosmo communicator | 16:31 |
| szopin | pie | 16:31 |
| szopin | found this and it mentions pie and exactly that namespace problem https://github.com/mer-hybris/hadk-faq/pull/4/commits/1cbbf5d2ba65a57f52a742987067d3d43701da37#diff-7b3ed02bc73dc06b7db906cf97aa91dec2b2eb21f2d92bc5caa761df5bbc168fR240 | 16:32 |
| szopin | not sure why it's not in the faq (or maybe there is pie specific faq) | 16:32 |
| mal | szopin: yes, you will need that | 16:34 |
| mal | I forgot to merge that | 16:34 |
| szopin | having problem installing through telnet Fatal error: droid-config-cosmocom-sailfish-1-202011251347.noarch requires droid-config-cosmocom = 1-202011251347, but this requirement cannot be provided | 16:35 |
| szopin | tried to copy all rpms from droid-configs and rpm -U *.rpm but that also fails | 16:35 |
| szopin | what's the suggested solution to install rpms from telnet? zypper? | 16:35 |
| mal | do not use *.rpm | 16:36 |
| mal | check what you have installed on the device and only install the once already installed | 16:37 |
| mal | "zypper se droid-config" | 16:37 |
| szopin | ok I just manually installed -sailfish first ignoring its dependencies and the first one went through | 16:38 |
| szopin | time for another logcat race | 16:38 |
| szopin | it's different now but still complains about selinux https://privatebin.net/?52902d7d8183e609#G46zse3ZD8adx7ZZJBCbSHt5aEiSQuagg1ubGmiot9se | 16:44 |
| szopin | weird half of the initial errors are gone but still some left | 16:46 |
| szopin | I guess I need to add those extra paths | 16:46 |
| mal | do you have vndservicemanager.rc in /usr/libexec/droid-hybris/system/etc/init/ | 16:46 |
| mal | nope | 16:47 |
| mal | you need probably some fixed rc | 16:47 |
| wdehoog | @mal is it known the the first link in the channel info text does not work anymore? | 16:47 |
| mal | yes | 16:47 |
| mal | we should remove it, I will someone to do that | 16:48 |
| T42 | <elros34> you should have it according to your droid-config https://github.com/Sailfish-on-Cosmo-Communicator/droid-config-cosmocom/blob/master/sparse/usr/libexec/droid-hybris/system/etc/init/vndservicemanager.rc | 16:49 |
| T42 | <elros34> he* | 16:49 |
| T42 | <elros34> could it be that this https://github.com/Sailfish-on-Cosmo-Communicator/droid-config-cosmocom/blob/master/sparse/usr/bin/droid/mount_overlay_system.sh mess something up? | 16:58 |
| T42 | <elros34> szopin try this: https://github.com/mer-hybris/droid-config-sony-nile/blob/master/sparse/usr/libexec/droid-hybris/system/etc/init/vndservicemanager.rc#L2 instead your path in vndservicemanager | 17:01 |
| mal | yes, that is what I meant | 17:02 |
| mal | the one in that other repos looks odd | 17:02 |
| mal | such overlay should not be needed | 17:03 |
| wdehoog | with latest libhybris I have no icons for the bottom row on the top menu (selfie etc.) the buttons seem to work, take up space but are invisible | 17:03 |
| szopin_ | thx, on it | 17:04 |
| mal | wdehoog: anything in logs | 17:07 |
| szopin_ | seems to be the same | 17:07 |
| mal | szopin_: did you remove the overlay thing? | 17:08 |
| szopin_ | comparing vndservicemanager.rc... there is just extra line at the end: override | 17:09 |
| szopin_ | removing that overlay.sh | 17:09 |
| T42 | <elros34> it can't be the same because now it try to load library "/system/lib64/libselinux_stubs.so, after fix it should use /usr/libexec/droid-hybris/system/lib64/libselinux_stubs.so | 17:10 |
| T42 | <elros34> in logcat | 17:10 |
| mal | yeah | 17:10 |
| T42 | <elros34> I see you have vndservicemanagerrr and vndservicemanagerrc-old, not sure which one is used | 17:12 |
| szopin_ | service vndservicemanager /vendor/bin/vndservicemanager /dev/vndbinder | 17:13 |
| szopin_ | setenv LD_PRELOAD /usr/libexec/droid-hybris/system/lib64/libselinux_stubs.so | 17:13 |
| szopin_ | this is from the file I think it was patched already locally | 17:13 |
| szopin_ | that's hwservicemanager... | 17:14 |
| szopin_ | .rc has setenv LD_PRELOAD /usr/libexec/droid-hybris/system/lib64/libselinux_stubs.so, the -old has setenv LD_PRELOAD /system/lib64/libselinux_stubs.so | 17:15 |
| szopin_ | but as it doesn't have .rc in filename it shouldn't be processed? | 17:15 |
| szopin_ | do they get included? then I need to ostore backups somewhere else, let me remove it just in case | 17:16 |
| szopin_ | ok, removed, overlay thingie gone, going to try rebuilding configs | 17:16 |
| T42 | <elros34> why don't you edit files directly on device? | 17:18 |
| szopin_ | never really learned vi | 17:18 |
| szopin_ | and there's no ed | 17:18 |
| szopin_ | (and it's been ages since I even used those) | 17:18 |
| T42 | <elros34> ah, you can add to patterns vim:) | 17:19 |
| szopin_ | would nano work? | 17:19 |
| T42 | <elros34> it is in jolla repos | 17:19 |
| szopin_ | ok, that would speed up tweaking .rc files for sure | 17:19 |
| szopin_ | hmm with droid-hal-init working zypper is taking forever to fail on fetching repos | 17:21 |
| szopin_ | the updated patterns will need --mic rebuild right? | 17:23 |
| szopin_ | for nano to get included... or will it end up as mw? | 17:23 |
| T42 | <elros34> patterns are in droid-config so you only rebuild droid-config and then build image. For now just download nano from jolla repos in target or from openrepos and install it | 17:25 |
| szopin_ | nothing provides info needed by nano-2.4.2-0.armv7hl | 17:28 |
| szopin_ | managed to install the droid-config...rpms in the meanwhile, next mic will definitely include it from the start | 17:29 |
| szopin_ | time to race logcat again | 17:29 |
| szopin_ | i guess it's no longer a race with hal-init working | 17:31 |
| szopin_ | latest, start looks clean now https://privatebin.net/?3a81b08cb166b40f#45EmWpzueHTNPSQPsXcDYoN6Tpzh7xJgPRKbkULGFsGZ | 17:32 |
| szopin_ | on a side note opensuse is great, no problems connecting to it 30h+ of idle time, no UI freezes, just occassional networking hiccups, but fedora had those too and hexchat seems to handle them better than Konversation | 17:42 |
| szopin_ | also latest journal log https://privatebin.net/?dcab73398c9addf9#G9ttUyu22wTNfdCyuXuRgADfWFcnFsa6bRvjw7uWWaWb | 17:47 |
| szopin_ | failed state at lipstick.service | 17:47 |
| szopin_ | library "libpq_cust.so" not found | 17:48 |
| mal | show logcat | 17:53 |
| szopin_ | https://privatebin.net/?3a81b08cb166b40f#45EmWpzueHTNPSQPsXcDYoN6Tpzh7xJgPRKbkULGFsGZ | 17:57 |
| mal | the journal log is missing the early messages, you should grab output of "journalctl -b --no-pager" right after reboot to so nothing gets removed because many services are failing all the time | 18:04 |
| szopin_ | ok, one sec | 18:05 |
| szopin_ | fresh https://privatebin.net/?cf0ba8251a5bae9c#HQBRbqS4hs6QuWJmsADay7LSo3rJsA8Y1SNkgy4nACcm | 18:06 |
| mal | does this work "EGL_PLATFORM=hwcomposer test_hwcomposer" | 18:15 |
| mal | it seems many android side services still fail | 18:16 |
| szopin__ | not sure if it went through as got another disconnect: test_hwcomposer: test_common.cpp:370: HWComposer* create_hwcomposer_window(): Assertion `err == 0' failed. | 18:17 |
| mal | yes, that is because the android side services fail to start | 18:21 |
| szopin__ | hmm, there is the 3.2.1 working build, I could try to salvage some files from it as it wasn't based on the yggdrasil tree but some secret mtk bsp | 18:23 |
| szopin__ | and NotKit cannot share it, just not sure which files would be helpful (only the rootfs of course as no access to source) | 18:24 |
| mal | can you compare the /usr/libexec/droid-hybris/system/etc/ld.config.28.txt with the one you have in /system/etc/ld.config.28.txt | 18:24 |
| szopin__ | sure | 18:26 |
| szopin__ | it's not in the rootfs | 18:33 |
| mal | is there any ld.config file somewhere in /system or /vendor | 18:34 |
| szopin__ | nope | 18:35 |
| szopin__ | at least not in rootfs, maybe it gets copied over from /vendor or such? | 18:35 |
| szopin__ | let me check the telnet | 18:35 |
| szopin__ | https://privatebin.net/?2e46ce1ef57948a7#5SMefSjLf2hBEhp6Z37bDfqhJGtAQhotBvDZVVkoqLho | 18:37 |
| szopin__ | if it gets modified at boot can try to reflash 3.2.1 and rerun same from there | 18:39 |
| szopin__ | I do have a frankenstein build (NotKit's gitlab --mic with release=3.4 and some changes I don't remember with a kernel that even boots it, but couldn't reproduce later on) that even has working modules and can ping from it, just UI apps have no network | 18:44 |
| szopin__ | but abandoned such tinkering trying to get a proper build | 18:44 |
| szopin__ | his rpms are on obs if anything useful from config perspective can be gleaned from them http://repo.merproject.org/obs/nemo:/devel:/hw:/planet:/cosmocom/sailfish_latest_armv7hl/armv7hl/ | 18:50 |
| szopin__ | and the resulting workin 3.2.1 rootfs can be found here https://gitlab.com/sailfishos-porters-ci/cosmocom-ci/-/jobs/432617888 | 18:51 |
| szopin__ | but again, it's based on mtk bsp not volla source tree | 18:52 |
| szopin__ | so there might be changes needed than sedding 6763 to 6771 | 18:54 |
| szopin__ | bigger* | 18:54 |
| mal | there is that /system/etc/ld.config.28.txt so what am I missing? | 19:08 |
| mal | ah, yeah, we mount the new file over it | 19:09 |
| szopin__ | init.rc from 3.2.1 has # MTK-specific: set nonencrypted on mount_all instead to fix nvram daemon | 19:12 |
| szopin__ | my 3.4 of course doesn't have that, doubt NotKit typed that all in... is there some patches for MTK I'm missing? | 19:12 |
| szopin__ | maybe from volla phone? | 19:12 |
| T42 | <ankaos> hı. how install 3.4.0.24 versin? | 19:14 |
| T42 | <ankaos> ı dont found link | 19:14 |
| T42 | <elros34> szopin: just to be sure, you flashed android base same as you used when building sailfish? | 19:15 |
| szopin__ | hmm, hard question, the android base is on boot partition 1 and was updated since Feb when NotKit built his 3.2.1 | 19:15 |
| szopin__ | but the base I'm using is just latest what NotKit uploaded when I asked planet when they will release source to allow porting sfos as NotKit left planet | 19:16 |
| szopin__ | elros34: just to give a better overview: there are 3 boot partitions for systems and 2 recovery ones, first is always android, 2nd is gemian from which I'm dd'ing sfos boot imgs, third is sfos (sfos rootfs needs to be put in ./stowaways/sailfishos to boot into it), first recovery partition is some planet custom installer that can flash images from sdcard/resize the partitions for linux/android, second | 19:21 |
| szopin__ | recovery is supposedly borked (planet asked to not touch it) | 19:21 |
| szopin__ | I'm guessing the first boot partition with android also is responsible for /vendor /firmware etc updates | 19:22 |
| szopin__ | so would be the 'you flashed android base' | 19:22 |
| szopin__ | and last android update was in July (5 months since the working 3.2 build) | 19:23 |
| szopin__ | but that 3.2 image from Feb is still working so don't think any android updates broke it | 19:25 |
| szopin__ | (then again if any source files like .rc or similar changed maybe it's not buildable now) | 19:25 |
| szopin__ | I do know 3.2.1 has ueventd.6763.rc that is nowhere to be found in the source tree for volla and it is included in rootfs | 19:25 |
| szopin__ | can try to add it | 19:26 |
| szopin__ | ok showed up in droid-hal-cosmocom.log | 19:35 |
| szopin__ | oh, vim was already in patterns | 19:38 |
| mal | szopin_: ueventd.*.rc is sometimes with different name in device repo | 19:43 |
| mal | more important is to check there is .rc files in out/ | 19:43 |
| szopin_ | nothing like that gets included, there is init.6771.rc | 19:43 |
| szopin_ | which I also manually copied to rootfs (though 3.2 port doesn't have that) | 19:43 |
| szopin_ | (oh the init 6771.rc might have been initially called 6763.rc but Android.mk from ./device gets somehow skipped I think as it doesn't copy files that are specifically mentioned there (like fstab.6763.rc which had to also manually copy)) | 19:46 |
| szopin_ | *dstab.6763 | 19:47 |
| szopin_ | *fstab | 19:47 |
| szopin_ | NotKit said he'll have a look at it over weekend, fingers crossed as it seems we're getting nowhere | 19:51 |
| T42 | <ankaos> ı am install 3.4.0.24 target and tooling | 20:03 |
| szopin_ | adding ueventd.6763.rc seems to have changed nothing, just in case journal https://privatebin.net/?12ed0bdf8e2873af#AEhMFwA5MzXJzwoyjfEkATVGdP1TpsnT1xMnngtF1ctw and logcat https://privatebin.net/?b9ca35976f22ac73#38DEziLJjxzdxnX87xUuho5S4NsgvCQ9cTQG6o39y65X | 20:04 |
| szopin_ | ankaos: if you have latest target and tooling the pulseaudio issue should be gone? | 20:05 |
| szopin_ | IF you also export RELEASE=3.4.0.24 before --mic? | 20:05 |
| szopin_ | or are you trying to build 3.3? | 20:05 |
| T42 | <ankaos> szopin: pulseaudio problem-> ı look. RELEASE=3.4.0.24 mic O.o | 20:07 |
| T42 | <ankaos> ahhh | 20:07 |
| T42 | <ankaos> :( | 20:07 |
| szopin_ | ? | 20:07 |
| T42 | <ankaos> mic relase 3.3.0.16 | 20:07 |
| T42 | <ankaos> but now mic release 3.4.0.24 | 20:07 |
| szopin_ | yeah, you're trying to build an older release, try to do 'export RELEASE=3.4.0.24' before running --mic | 20:08 |
| szopin_ | or edit .ks if it has RELEASE defined | 20:08 |
| T42 | <ankaos> ı am install 3.4.0.24 target and tooling | 20:08 |
| szopin_ | good | 20:08 |
| T42 | <ankaos> now rebuild sfos | 20:09 |
| szopin_ | yup, just make sure RELEASE=3.4... | 20:09 |
| T42 | <ankaos> maybe fix pulseaudio probelm | 20:09 |
| T42 | <ankaos> what is this? -» The following 6 NEW packages are going to be installed: | 20:10 |
| T42 | <ankaos> bluez5 bluez5-libs bluez5-obexd droid-config-oxygen-bluez5 kf5bluezqt-bluez5 kf5bluezqt-bluez5-declarative | 20:10 |
| T42 | <ankaos> The following 6 packages are going to be REMOVED: | 20:10 |
| T42 | <ankaos> bluez bluez-configs-mer bluez-libs kf5bluezqt-bluez4 kf5bluezqt-bluez4-declarative obexd | 20:10 |
| szopin_ | if it breaks your build try to remove bluez4 from your sb2 | 20:10 |
| szopin_ | as you're building bluez5 it seems | 20:11 |
| T42 | <ankaos> ı am waiting build | 20:11 |
| szopin_ | should be fine, the only bluez things I have in history is: sb2 -t $VENDOR-$DEVICE-$PORT_ARCH -m sdk-install -R zypper -n rm bluez5-configs-mer | 20:18 |
| szopin_ | sb2 -t $VENDOR-$DEVICE-$PORT_ARCH -R -msdk-install zypper in bluez5 | 20:18 |
| szopin_ | so you can try those if it fails | 20:18 |
| T42 | <ankaos> okey. ı am waiting build | 20:20 |
| szopin_ | sounds good, if it takes a while then you're behind the last hurdle | 20:21 |
| xmn | Hey let me ask you guys. I might be getting great battery life under megi multiboot image of sailfish 3.3.0.16. Where as the 3.4 vers runs warm to the touch and battery last only out 5+- hours in airplane mode and no apps running. | 20:31 |
| szopin | any processes running under top/htop? | 20:31 |
| szopin | some people have reported web browser not failing right, if it's hanging at 100% core usage sounds like it could be it | 20:32 |
| xmn | ic, testing the battery under th multboot at the moment so can't check right now. | 20:33 |
| xmn | but good idea to check that | 20:33 |
| xmn | I was guessing that the upgrade to3.4 I did remove or wrote over some of the pineohone tweaks that might have helped | 20:33 |
| xmn | so I was thinking if I could load megi kernal into 3.4 version I have. | 20:34 |
| szopin_ | godammit those reconnects, but anyway, if it's warm it sounds like some runaway process going 99-100% cpu usage | 20:37 |
| szopin_ | browser would fit the bill and it is now used not only by browser but also email client | 20:37 |
| szopin_ | install lighthouse from openrepos or have htop running next time | 20:37 |
| szopin_ | specific pinephone tweaks... maybe, if it no longer goes into sleep/idle, but that shouldn't go down to 5h (then again there is pureOS with similar uptime) | 20:40 |
| szopin_ | gemini on cosmo with no deep sleep still will easily clock in 12h or so | 20:42 |
| xmn | oh wow | 20:42 |
| szopin_ | gemian* | 20:42 |
| xmn | yeah using light house and systemdatascope | 20:43 |
| szopin_ | yeah, it is always alive as a normal debian distro, but with no processes running in the bg it doesn't drain nowhere near that fast | 20:43 |
| xmn | will for sure look into the browser once back on the other install | 20:43 |
| szopin_ | just check any process, browser was reported by some people on 3.4 as not closing correctly, but it might be anything really | 20:47 |
| szopin_ | if it's using one core fully it would explain warmth and the drain | 20:47 |
| xmn | makes sense | 20:48 |
| xmn | this current multi boot is cool to the touch in resting state | 20:48 |
| szopin_ | multi-boot means you can choose what to boot, you're not booting multiple oses at the same time right? | 20:49 |
| szopin_ | so booting into pmos or ubports has nothing to do with sfos afaik | 20:49 |
| szopin_ | the temp of different oses is just their base state | 20:49 |
| xmn | correct | 20:50 |
| xmn | but does contain some custom megi tweaks I think. Just not sure how much. | 20:50 |
| xmn | but each OS has it own btfs space | 20:51 |
| xmn | it very handy for testing | 20:51 |
| szopin_ | no idea what megi is, cosmo is multiboot, android and sfos (and ubports presumably) will easily clock in 48h in passive flight mode state, gemian maybe 18 | 20:51 |
| szopin_ | not sure what btfs space is | 20:52 |
| szopin_ | are they maybe using btrfs snaps? | 20:53 |
| xmn | apologies, https://xnux.eu/p-boot-demo/ | 20:53 |
| xmn | from what I understand the bfts space allow you to revert back to the orig image with out flashing the card again. | 20:54 |
| szopin_ | yeah it's just btrfs snapshotting | 20:55 |
| xmn | yup | 20:55 |
| szopin_ | would call it brave on a phone but jolla did it on j1 | 20:56 |
| szopin_ | (and it wasn't pretty) | 20:56 |
| xmn | I'm ignorant to many of these things :) | 20:56 |
| xmn | lol | 20:56 |
| szopin_ | I gave up on btrfs on my 30GB of SSD space, safer on ext4 or maybe just afraid of change | 20:57 |
| szopin_ | or maybe just had j1 and seen how it can fsck up things | 20:57 |
| szopin_ | (opensuse forums also discouraged people from using btrfs on <128GB then again full desktop distro so maybe I'm biased) | 20:58 |
| szopin_ | but just in case, my understanding is btrfs snaps is just having a backup copy of your previous settings, nothing to do with 'cool to the touch' whatever you're booting atm will be responsible for the temp, so if sfos is warm some process is going crazy, what you have installed beside it has no impact | 20:59 |
| xmn | yup, never thought bfts was in anyway impacting the temp. | 21:01 |
| szopin_ | I meant: this current multi boot is cool to the touch in resting state | 21:01 |
| xmn | Yeah I'm not running btfs on anything. I'm not experienced enough to get out of that trouble :0 | 21:01 |
| szopin_ | whatever is in your multiboot is just resting on the disk | 21:01 |
| xmn | yeah | 21:01 |
| szopin_ | with enough disk space btrfs should be a blessing (jolla adopted it very early like what 8 years ago?) | 21:02 |
| szopin_ | but the curse of it only then became clear | 21:02 |
| xmn | heh | 21:03 |
| szopin_ | I don't think btrfs is used in any of the latest ports... 8 years later | 21:03 |
| xmn | maybe in another 5 years :) | 21:03 |
| szopin_ | probably same with wayland | 21:03 |
| szopin_ | even on fedora and opensuse wayland is still not properly working out of the box with hdmi or similar | 21:03 |
| szopin_ | with generic nvidia cards on laptops | 21:03 |
| szopin_ | 8 years after jolla adopted it... then again for mobile phones it kinda works great except the hdmi output | 21:04 |
| szopin_ | it's absurd to expect year of the linux desktop when we're in 2020, 13 years after I had ubuntu 7.04 running on a laptop, biggest issue then was suspend, it was killing laptops left and right, 13 years later linux (fedora/opensuse/ubuntu... tested them all and all have same issues) can't provide proper hdmi output | 21:07 |
| szopin_ | yeah 5 extra years seems right | 21:08 |
| szopin_ | xmn: btw how do you rate OS's on pinephone? | 21:14 |
| xmn | yeah sometime I feel like if some project worked better together to at least get one thing fully polished then move on and help the next thing | 21:30 |
| xmn | szopin_ I have a bias to debian based since it what I learned via n900 and my server. | 21:30 |
| xmn | But overall I'm liking the progress of manjaro very much | 21:31 |
| xmn | maemo shows promise. sailfish is so nice if I could jurt get rid of a few issue it could be a daily. Mobian is nice for apps, but is moving seemly very slow, so maybe in the future. So looking at long term testing of sailfish, mobian, manjaro and maemo-leste. | 21:33 |
| xmn | But who could image that webos has a lot of stuff already dialed in :) | 21:33 |
| xmn | Not really feeling kde or ubport right now either | 21:34 |
| xmn | pretty happy and thankful for the progress so far. | 21:34 |
| T42 | <ankaos> Info[11/26 21:34:07] : marking pattern jolla-configuration-oxygen 1-202011262107 to be installed | 21:40 |
| T42 | <ankaos> Warning[11/26 21:34:07] : repo problem: nothing provides pulseaudio >= 12.2+git8 needed by pulseaudio-modules-droid-hidl-1.3.0+master.20200525095923.a35a5ba-1.armv7hl | 21:40 |
| T42 | <ankaos> https://github.com/enesutku07/pulseaudio-modules-droid-hidl edit | 21:44 |
| T42 | <ankaos> PlatformSDK enes@enes:~/hadk$ sudo rm -rf /var/tmp/mic/cache/ | 21:49 |
| T42 | <ankaos> adaptation-community-common-oxygen-3.3.0.16/ etc/ solv/ | 21:49 |
| T42 | <ankaos> adaptation-community-oxygen-3.3.0.16/ hotfixes-3.3.0.16/ var/ | 21:49 |
| T42 | <ankaos> apps-3.3.0.16/ jolla-3.3.0.16/ | 21:49 |
| T42 | <ankaos> customer-jolla-3.3.0.16/ raw/ | 21:49 |
| T42 | <ankaos> this is normal? | 21:49 |
| mal | seems reasonable, why do you ask? | 21:49 |
| mal | ah, did you rebuild package using build_packages.sh? | 21:49 |
| mal | and did you really set correct RELEASE before running mic? | 21:50 |
| T42 | <ankaos> no | 21:50 |
| mal | hadk pdf says "export RELEASE=3.3.0.16" which you of course need to change to be correct version | 21:51 |
| mal | before running mic in chapter 8.4 | 21:51 |
| T42 | <ankaos> ı dont use chapter 8.4 | 21:51 |
| mal | what do you use then? | 21:52 |
| T42 | <ankaos> only ı use rpm/dhd/helpers/build_packages.sh | 21:52 |
| mal | if rad chapter 8.4 it says "rpm/dhd/helpers/build_packages.sh --mic" | 21:52 |
| T42 | <adampigg> mal, the new pinephone wlan driver exposes 2 interface, wlan0 and wlan1 ... wlan0 should be default, wlan1 is for p2p i think ... how do i tell connman / wpa_supplication to use wlan0 | 22:13 |
| T42 | <adampigg> Nov 26 22:06:22 PinePhone wpa_supplicant[3260]: wlan1: CTRL-EVENT-STARTED-CHANNEL-SWITCH freq=2437 ht_enabled=1 ch_offset=-1 ch_width=40 MHz cf1=2427 cf2=0 | 22:13 |
| T42 | <adampigg> Nov 26 22:06:24 PinePhone wpa_supplicant[3260]: wlan1: CTRL-EVENT-ASSOC-REJECT status_code=1 | 22:13 |
| T42 | <adampigg> Nov 26 22:06:24 PinePhone wpa_supplicant[3260]: wlan1: CTRL-EVENT-SSID-TEMP-DISABLED id=0 ssid="pgznet" auth_failures=1 duration=10 reason=CONN_FAILED | 22:13 |
| T42 | <elros34> maybe check whether your wlan modules support parameters then you could load it like: modprobe wlan ifname=wlan0 if2name=p2p0 | 22:18 |
| T42 | <adampigg> Ok | 22:20 |
| T42 | <ankaos> "Build pulseaudio-modules-droid-hidl? [Y/n/all]n" working build | 22:31 |
| T42 | <ankaos> mal and szopin: working build | 22:34 |
| Umeaboy | Hi! | 22:43 |
| T42 | <ankaos> Umeaboy: hı | 22:44 |
| Umeaboy | Will there be an LOS-17.1 patches branch anytime soon for newer devices made by Samsung? I've been looking to build the AOSP as Lineage fails to build due to lack of memory regardless how I use the JACK server? | 22:47 |
| Umeaboy | I am running an unofficial build of Lineage already. | 22:47 |
| T42 | <Amy Christy Sweeney %lastname%> I can't explain how l feel each time I receive my profits from this company. What can I say? Thanks to Mr Tomas, I've made so much in 2 weeks @Profits_world100 | 23:20 |
| Umeaboy | Money isn't everything when you loose yourself inside because of it. | 23:27 |
| T42 | <edp_17> @elros34 : Interesting as I have found the following messages when the external tv was connected (journalctl): "Blanking external display", "Unblanking internal display" and loads of "Waiting and closing retire fence fd" | 23:56 |
| T42 | <edp_17> @elros34 : I have two different cables and tried both. Some logs are here: cable1: https://paste.ubuntu.com/p/p76HkMj9vM/ and cable2: https://paste.ubuntu.com/p/dpXVRx9yhh/ | 23:57 |
| T42 | <edp_17> It looks like your modified qt5-qpa-hwcomposer-plugin is reporting into journal but the screen doesn't appear on external display. | 23:58 |
| T42 | <elros34> what libhybris version do you use? | 23:58 |
Generated by irclog2html.py 2.17.1 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!
