| Mister_Magister | mal: https://build.sailfishos.org/project/show/nemo:testing:hw:motorola:titan 5.0 built, wpa_supplicant as well | 08:47 |
|---|---|---|
| Mister_Magister | in case you want to verify it | 08:47 |
| Mister_Magister | i don't remember how you update to 5.0 lmao | 09:19 |
| Mister_Magister | repos like storman or chum fail cause they don't have 5.0.0.x repos | 09:21 |
| Mister_Magister | even adaptation-community-common is failing xd | 09:22 |
| Mister_Magister | I can't figure out how to update to 5.0 | 09:35 |
| Mister_Magister | ohhhh .72 is not yet on community common | 09:36 |
| Mister_Magister | updating now | 09:46 |
| Mister_Magister | lets see | 10:00 |
| Mister_Magister | mal: what's supposed to not work? wpa_supplicant so wifi? if so then it's working, in fact I think everything is working fine | 10:02 |
| Mister_Magister | hmm | 10:05 |
| Mister_Magister | after updating adaptation-community switched repo to 5.0 | 10:05 |
| Mister_Magister | so why do we need specific versions then xd | 10:05 |
| Mister_Magister | i'm building 5.0.0.67 image for all of my ports on jenkins rn | 10:17 |
| Mister_Magister | looong pipeline | 10:18 |
| Mister_Magister | okay sensors are dead | 10:28 |
| Mister_Magister | let's see if turning it off and back on again fixes it | 10:29 |
| Mister_Magister | nope, they're dead alright | 10:31 |
| Mister_Magister | ok for some reason it installed hybris-libsensorfw-qt5-binder instead of hybris-libsensorfw-qt5-hal | 10:35 |
| Mister_Magister | I've changed hybris-libsensorfw-qt5 to hybris-libsensorfw-qt5-hal in patterns | 10:37 |
| mal | you need to force correct package in spec | 11:17 |
| mal | same for all of those *-hal packages | 11:17 |
| mal | I mean in patterns of course | 11:20 |
| Mister_Magister | gotcha | 11:37 |
| Mister_Magister | i mean beside libsensorfw there isn't anything else | 11:37 |
| mal | Mister_Magister: geoclue plugin is another one which might cause issues | 11:41 |
| Mister_Magister | thanks, but ye, wifi works just fine so thank you so much for the fix and for keeping half of my ports alive <3 | 11:42 |
| Mister_Magister | longest supported sfos device is still supported | 11:42 |
| mal | good | 11:42 |
| mal | so old devices should use geoclue-provider-hybris-hal, check your patterns | 11:42 |
| Mister_Magister | i | 11:44 |
| Mister_Magister | will | 11:44 |
| mal | I'll try to get one more bugfix release from 5.0 with that wpa_supplicant fix and probably some other fixes | 11:44 |
| Mister_Magister | we had 5.0 for… like ayear | 11:47 |
| Mister_Magister | geoclue was incorrect so thanks for the tip | 11:49 |
| Mister_Magister | mal: on my device i ONLY have geoclue-provider-hybris-hal | 12:16 |
| Mister_Magister | ah no sorry i'm lying and blind | 12:16 |
| *** xmn_ is now known as xmn__ | 12:28 | |
| *** xmn__ is now known as xmn_ | 12:28 | |
| Mister_Magister | mal: it would seem that GPS is not working | 12:29 |
| Mister_Magister | i have the icon but both gpsinfo and CSD fail to get any sattelite info, not even compass | 12:29 |
| Mister_Magister | I've updated provider-hybris to 0.2.38 | 12:31 |
| Mister_Magister | geoclue-master and geoclue-hybris seem to be running | 12:33 |
| Mister_Magister | I'm getting a lot of these in logcat when geoclue-hybris starts https://paste.opensuse.org/pastes/edbfee1eee40 | 12:36 |
| *** xmn_ is now known as xmn | 12:37 | |
| Mister_Magister | when i added QT_LOGGING_RULES="*.*=true" it seems to be doing something so maybe it's not communicating idk | 12:38 |
| Mister_Magister | in gpsinfo there's no sattelite view which usually indicates something is broken xd | 12:39 |
| Mister_Magister | https://paste.opensuse.org/pastes/8145ce0b7dc6 it's just spamming this | 12:40 |
| Mister_Magister | wait i think i remembered something from 2016 | 12:42 |
| Mister_Magister | nope that wasn't it | 12:43 |
| Mister_Magister | when there's no compass it usually indicates some communication issue | 12:44 |
| Mister_Magister | ideas welcome xd | 12:58 |
| Mister_Magister | mal: on moto g2 i don't have gps but on x2 I do so that's weird | 14:46 |
| mal | didn't you have some odd issues like that in the past also? | 14:50 |
| Mister_Magister | i did, but i tired the usual fix and it didn't help | 14:50 |
| Mister_Magister | usual fix is to erase two partitions but that wasn't it sadly | 14:51 |
| Mister_Magister | i can flash 4.6 for testing though | 14:51 |
| Mister_Magister | i'm updating second g2 device, if it won't work still then it 100% will be an issue | 14:57 |
| Mister_Magister | mal: okay 2014 g2 gps works fine | 15:23 |
| Mister_Magister | so I guess it's non issue | 15:24 |
| Mister_Magister | mal: aaand guess I was impatient, after i froze to death outside the sattelites started popping up | 15:26 |
| Mister_Magister | so ig it works on all 3 devices | 15:26 |
| Mister_Magister | and also your fix works on all 3 distinct devices (g2 2014 ver, g2 2015 ver, x2) | 15:27 |
| mal | good, it was approved already also so I will merge it soon | 15:28 |
| Mister_Magister | sweet | 15:28 |
| Mister_Magister | thanks again for keeping my devices alive <3 | 15:28 |
| mal | I'll try to keep systemd also working, latest one will need some reverting to make it work on old devices | 15:31 |
| Mister_Magister | mal: oh I noticed soemthing, my x2 lost 4G option | 15:31 |
| mal | does it support that? | 15:31 |
| mal | can you show your config repo? | 15:32 |
| Mister_Magister | mal: it did on 4.6 | 15:33 |
| Mister_Magister | https://github.com/VerdandiTeam/droid-config-victara/ | 15:33 |
| Mister_Magister | oh | 15:35 |
| Mister_Magister | it hs enable4G=false | 15:35 |
| Mister_Magister | why in the world | 15:35 |
| Mister_Magister | and i have 4G back, dummy me | 15:38 |
| Mister_Magister | all 3 devices updates released | 16:04 |
| Mister_Magister | i'm free xd | 16:04 |
| Mister_Magister | it's quite a lot of building and typing and uploading even though most of it is automated | 16:04 |
| joff | hi all, someone know how work bluetooth for ruuvi devices or car ? | 16:38 |
| joff | i have this log for ruuvi basic bluetooth connection from bluetoothctl | 16:39 |
| joff | https://debian-facile.org/paste-C8825CB1B6-b4fa87cbc41c2fd571a5b905cf5d8ba8 | 16:39 |
| joff | O dear ! App function skruuvi with the device ! | 16:40 |
| Keto | joff: #sailfishos is better place for these end-user questions, and you don't need to post them on both channels. This channel is for Sailfish OS porting. | 17:12 |
| joff | yes, sorry | 17:12 |
| Mister_Magister | mal: do you have 5 seconds to help me out with lipstick restart? | 18:47 |
| mal | what is the issue? | 18:53 |
| Mister_Magister | mal: on op6 when i restart lipstick, it doesn't work second time, like black screen, nothing | 18:54 |
| Mister_Magister | oh sorry, correctoind, it just shows what was in the buffer | 18:55 |
| Mister_Magister | people want to use patch manager and that kinda stops them | 18:55 |
| mal | do you have proper hwc restart config? new devices need that | 18:58 |
| mal | well most with a8+ or something | 18:58 |
| Mister_Magister | i have no clue i never heard about it | 18:59 |
| Mister_Magister | mine is lineage16 based | 18:59 |
| Mister_Magister | which would be 9 | 18:59 |
| Mister_Magister | here's https://github.com/VerdandiTeam/droid-config-enchilada/ if it's needed | 18:59 |
| Mister_Magister | could you point me to that hwc restart config? | 19:00 |
| mal | you find out what hwc service is called in android .rc then add a config | 19:00 |
| mal | trying to give example but github is crashing | 19:00 |
| Mister_Magister | /vendor/bin/hw/android.hardware.graphics.composer@2.1-service is running | 19:01 |
| mal | yes, but the name of the service, not the binary name | 19:01 |
| mal | so grep for that filename in /vendor/etc/init or something | 19:02 |
| Mister_Magister | it's the executable that's running | 19:02 |
| Mister_Magister | service vendor.hwcomposer-2-1 /vendor/bin/hw/android.hardware.graphics.composer@2.1-service | 19:02 |
| Mister_Magister | in /vendor/etc/init/android.hardware.graphics.composer@2.1-service.rc | 19:02 |
| Mister_Magister | oh ye, killing it and restarting lipstick, did work | 19:03 |
| Mister_Magister | so it's probably that | 19:03 |
| mal | https://pastebin.com/4fyWBya2 as /etc/mce/60-compositor-enchilada.ini | 19:04 |
| Mister_Magister | thank you kindly | 19:04 |
| mal | that works since sfos 4.6 | 19:04 |
| Mister_Magister | i'm on 5.0 already | 19:05 |
| Mister_Magister | that worked thank you! | 19:08 |
| Mister_Magister | lol github works fine for me but now OBS can't clone repo xd | 19:13 |
| mal | github has been having a lot of issue today https://www.githubstatus.com/ | 19:15 |
| Mister_Magister | mal: if you're bored, would you mind looking at two patches someone vibe coded? I HATE to admit it but they did fix volte for me (it was previously dropping to 2G when trying to call) | 19:25 |
| Mister_Magister | https://jottacloud.com/s/058fdc812f925c041368c455feea24e2d5a/thumbs/install_package/patches | 19:25 |
| Mister_Magister | what's funny is that on o6t volte works but on op6 it doesn't. And you'll ask "well what's the difference between thes…" NONE. same cpu, same kernel, same everything beside different notch, in screen fingerprint, lack of 3.5mm and notification led | 19:35 |
| Mister_Magister | so i don't understand why doesn't it work | 19:35 |
| mal | need to think about those | 19:47 |
| mal | those qti plugin changes seem quite simple, maybe you should make a PR to that project, it's not something I'm involved in any way | 19:48 |
| mal | but those binder plugin changes need more thinking | 19:49 |
| Mister_Magister | Ill test if one or both patches are needed | 19:49 |
| Mister_Magister | ad the ext-qti one is not doing much | 19:49 |
| mal | so you were trying to fix the drop to 2g issue or also something else? | 19:53 |
| Mister_Magister | well it's not my code, i don't use "AI", just someone on forum posted these and said its working. I just built binder-plugin and binder-plugin-ext-qti, and on op6t i had it register (also after doing weird flashing) but it was dropping to 2G and got stuck as neither of us had any idea how to debug it | 19:54 |
| mal | if so maybe also QTI_RADIO_CALL_DOMAIN_AUTOMATIC could be an option for that? or whatever that is called in that plugin | 19:54 |
| Mister_Magister | I can do some more testing sure | 19:55 |
| mal | so did those fix op6 registering also? | 19:55 |
| Mister_Magister | no op6 is not registering | 19:55 |
| Mister_Magister | at all | 19:55 |
| mal | ok | 19:55 |
| Mister_Magister | i did the same process on op6 and it just doesn't work | 19:56 |
| mal | ok, maybe test only that domain change and see what happens | 19:56 |
| mal | or op6t | 19:56 |
| Mister_Magister | could you explain that change as i don't get what do you mean exactly | 19:57 |
| Mister_Magister | let me find the info about flashing the thingy so that i remember what was it about | 19:58 |
| Mister_Magister | ah it was flashing some partition with specific MBN | 19:59 |
| mal | I mean original code has: dial_request_writer->call_details.call_domain = QTI_RADIO_CALL_DOMAIN_UNKNOWN; and dial_request_writer->has_call_details = FALSE; and new change first to PS and second to TRUE, maybe automatic might be better than PS | 20:04 |
| Mister_Magister | ah gotcha | 20:07 |
| Mister_Magister | right now i'm building the upstream version to see if it really doesn't work | 20:07 |
| Mister_Magister | maybe it magically fixed itself | 20:07 |
| mal | I wonder if it even needs call details | 20:07 |
| Mister_Magister | I'll do some testing and let you know | 20:09 |
| Mister_Magister | thanks for having a look | 20:10 |
| Mister_Magister | as to op6, maybe volte will work on mainline xd | 20:11 |
| Mister_Magister | op6 is kind of mainline go-to device right now so i plan to make mainline based port | 20:12 |
| Mister_Magister | yeah no it instantly drops to 2G on upstream version | 20:14 |
| Mister_Magister | so there is something | 20:14 |
| mal | so try all of the changes in volte plugin but not in binder plugin | 20:18 |
| mal | which one did you try? | 20:18 |
| Mister_Magister | yeah like i said i'll try then on each own | 20:18 |
| Mister_Magister | slow down for now i only tired upstream xd | 20:19 |
| Mister_Magister | ok binder alone still drops | 20:19 |
| Mister_Magister | okay -ext changes make it not drop | 20:22 |
| Mister_Magister | so binder changes are completely optional | 20:22 |
| Mister_Magister | changed the QTI_RADIO_CALL_DOMAIN_AUTOMATIC and will let you know | 20:25 |
| Mister_Magister | mal: you were spot on, automatic did the trick | 20:27 |
| Mister_Magister | yeah works | 20:27 |
| Mister_Magister | wait sorry | 20:28 |
| Mister_Magister | i tested wrong branch | 20:29 |
| Mister_Magister | also mal there's no dialing sound is that normal? | 20:29 |
| mal | you mean the ringback tone? | 20:29 |
| Mister_Magister | i don't know what i mean, the dailing tone like when you're dailing | 20:30 |
| mal | those sounds that should be there when waiting for other end to answer? | 20:30 |
| Mister_Magister | yes | 20:30 |
| Mister_Magister | also update, the AUTOMATIC did not do the trick | 20:30 |
| Mister_Magister | but it works with the AI patch | 20:31 |
| mal | what other changes did you have with automatic? | 20:32 |
| Mister_Magister | just the automatic | 20:32 |
| mal | try doing also that dial_request_writer->has_call_details = TRUE; | 20:33 |
| Mister_Magister | https://github.com/VerdandiTeam/ofono-binder-plugin-ext-qti/commit/db2efebcfab4098d13992e50a1c09be7bd8f4083 | 20:33 |
| Mister_Magister | heres ze commit | 20:33 |
| Mister_Magister | mal: still drops | 20:37 |
| mal | does it work with the full patch but with automatic? | 20:37 |
| Mister_Magister | i shall try that next | 20:38 |
| Mister_Magister | oh sorry i'm dumb i didn't commit | 20:38 |
| Mister_Magister | if it doesn't work after actually testing with call details i'll test with full patch and let you know | 20:39 |
| Mister_Magister | ye okay it works with has_call_details | 20:40 |
| Mister_Magister | confirmed with full call it works fine i can call and have 4G mobile data so it is 4G | 20:42 |
| Mister_Magister | I wonder if it works with none and just call details | 20:44 |
| mal | try that also | 20:45 |
| mal | but automatic seems like the logical value since that should allow fallback to 2g if needed | 20:46 |
| mal | not sure how that PS works in bad cases | 20:46 |
| T42 | <Mister_Magister> it's already falling back to 2G | 20:46 |
| Mister_Magister | that's the issue | 20:46 |
| mal | yes but there are also normal situations where fallback should happen, of course not always but in service strength is bad | 20:47 |
| mal | *if | 20:47 |
| Mister_Magister | mal: i just got info from one user that it works on op6 | 20:47 |
| Mister_Magister | so maybe my operator doesn't like op6 | 20:48 |
| mal | yeah | 20:48 |
| Mister_Magister | and it's nice that MBN flashing is not needed | 20:49 |
| Mister_Magister | mal: yeap call_details is all that's needed | 20:51 |
| Mister_Magister | unless my sleepy brain forgot to commit again | 20:51 |
| Mister_Magister | no it's correct commit | 20:51 |
| mal | hmm, interesting | 20:56 |
| Mister_Magister | so we boiled down AI to one line, good | 20:57 |
| mal | maybe some devices are somehow different | 20:57 |
| mal | but I think the automatic value might be nicer though | 20:58 |
| Mister_Magister | not to sound pedantic but different devices being… well… different? no… can't be :D | 20:58 |
| mal | just odd why some devices don't need call details | 20:58 |
| Mister_Magister | I'll for now leave it to use my fork and also switch droid-config to use it | 21:00 |
| Mister_Magister | mal: quick question, settings show prefer 5G option, how can i disable it (my phone has no 5G) | 21:01 |
| mal | if you have new enough binder-plugin you can use https://github.com/mer-hybris/s19mps-droid-configs/blob/master/sparse/etc/ofono/binder.d/s19mps.conf#L5 | 21:02 |
| mal | that last line, and first of course to make the config valid | 21:03 |
| T42 | <Mister_Magister> that did the trick thanks | 21:06 |
| Mister_Magister | oh also you never answered, how about the dailing tone? can it be fixed? xd | 21:09 |
| mal | it can, but it's not implemented in that plugin yet, RINGBACK_TONE_INDICATION | 21:28 |
| T42 | <Mister_Magister> ahh | 21:30 |
| mal | and relevant part in ofono-binder-plugin https://github.com/mer-hybris/ofono-binder-plugin/blob/master/lib/include/binder_ext_call_impl.h#L84 | 21:30 |
| T42 | <Mister_Magister> so maybe in the future | 21:31 |
| mal | official volte plugin has it | 21:31 |
| mal | it's quite simple to implement, the indication binder message only contains a boolean | 21:34 |
| Mister_Magister | ye but that'd require… effort | 21:36 |
| mal | heh | 21:37 |
| Mister_Magister | also its simple when you know what to do | 21:37 |
| mal | gbinder is not that difficult when simple values are used | 21:38 |
| mal | also the volte plugin code probably has good examples how to do the rest i.e. the singal handling and how to send info back to binder plugin | 21:39 |
| Mister_Magister | mal: do I need the StartHwCompositor ? | 22:02 |
| Mister_Magister | i'll remove it cause it causes gui to not work on boot sometimes | 22:33 |
| mal | hmm, strange | 22:35 |
| Mister_Magister | nah i lied it didn't help | 23:25 |
| Mister_Magister | for just one guy, it causes ui not to start | 23:25 |
Generated by irclog2html.py 2.17.1 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!