| T42 | <elros34> @the_hiktor script should use one .config file related to kernel but it finds also some other dtb relate .config fles. Hardcode it here https://github.com/mer-hybris/droid-hal-device/blob/master/droid-hal-device.inc#L472. Know issue but nobody with this issue bother to fix it | 04:21 |
|---|---|---|
| T42 | <the_hiktor> this is the part you are talkng about. right? | 07:23 |
| T42 | <the_hiktor> https://hastebin.com/share/vosawanuja.bash | 07:23 |
| Mister_Magister | mal: Hiiii if you're free would you mind giving me a hand? | 09:43 |
| T42 | <frullewulle> which branch of hybris-boot is the best to use with hybris-22.2 ? a16 ? | 10:32 |
| T42 | <the_hiktor> elros, i defind the exact path for the kernel .conf | 10:35 |
| T42 | <the_hiktor> it worked but faced this | 10:36 |
| T42 | <the_hiktor> https://hastebin.com/share/uvevuxojic.bash | 10:36 |
| mal | just a reminder that the latest sailfish release is 5.0.0.77 | 11:28 |
| mal | when building images that should be used, not .67 like I see being used | 11:29 |
| Mister_Magister | i didn't even update my ports to .77 lol | 11:59 |
| Mister_Magister | so mal any ideas or are you busy still? | 12:07 |
| Mister_Magister | i'm just asking because when you said "does the device use many modules" sounded like "aha!" moment :D | 12:10 |
| mal | well if the module loading fails then it could cause all kinds of issues | 12:47 |
| mal | I assume you flashed also vendor_boot.img when you flashed hybris-boot.img | 12:48 |
| mal | from the same build | 12:48 |
| Mister_Magister | venor_boot and dtbo yes | 12:49 |
| Mister_Magister | i copied the flashing stuff from my other device which also needed flashing vendor_boot and dtbo | 12:49 |
| mal | you saw nothing in host dmesg? | 12:50 |
| Mister_Magister | nothing at all | 12:50 |
| mal | and the boot images were built after adding the missing boot partition to fstab? | 12:51 |
| mal | and fixup worked for that? | 12:51 |
| Mister_Magister | yes and yes | 12:52 |
| Mister_Magister | i'll try backing off some kernel config changes, just do the errors and see | 13:00 |
| Mister_Magister | mal: https://paste.opensuse.org/pastes/a0ec5d27eddc | 13:01 |
| Mister_Magister | it worked fine | 13:01 |
| mal | now it booted or what do you mean? | 13:10 |
| T42 | <Mister_Magister> no the log, the fixup worked fine | 13:16 |
| *** Mister_Magister_ is now known as Mister_Magister | 13:37 | |
| Mister_Magister | mal: i thought of checking /data if init.log is there, so that i know if its in rootfs or fails in ramdisk AND I GOT LOG! | 15:13 |
| Mister_Magister | https://paste.opensuse.org/a0bd4e5f481c | 15:13 |
| Mister_Magister | mkdir: can't create directory '/config/usb_gadget/g1/functions/rndis.usb0': No such file or directory yeah as expected | 15:13 |
| Mister_Magister | i thought kernel config might be missing | 15:14 |
| Mister_Magister | and then it reboots after minute because it couldn't set up usb from the look of it? | 15:14 |
| Mister_Magister | I remember adding some kernel config about g_gadget… but can't put my finger on it | 15:16 |
| Mister_Magister | # CONFIG_USB_CONFIGFS_RNDIS is not set yeah that'd do it | 15:20 |
| Mister_Magister | now just gotta wait an hour for it to build kernel | 15:21 |
| Mister_Magister | ayy init-debug in real rootfs! | 16:22 |
| Mister_Magister | and then it disconnected and no rndis | 16:24 |
| Mister_Magister | i can try masking usb-moded | 16:24 |
| Mister_Magister | ayy i'm in | 16:29 |
| Mister_Magister | mal: /vendor seems to be erofs? i don't think sfos has that? | 16:32 |
| mal | hmm | 16:33 |
| T42 | <elros34> @the_hiktor so where do you have this libdl.so and other missing libs (use find command) | 16:35 |
| Mister_Magister | i'll try to compile erofs-utils on sfos obs | 16:37 |
| mal | util-linux has some erofs things | 16:39 |
| mal | what does mount say if you try to mount it? | 16:40 |
| Mister_Magister | ext4 | 16:40 |
| Mister_Magister | huh | 16:40 |
| Mister_Magister | then fstab was incorrect i guess? | 16:41 |
| Mister_Magister | when i didn't specify erofs it worked like a charm | 16:41 |
| mal | hmm | 16:41 |
| T42 | <the_hiktor> @elros34 here https://hastebin.com/share/wijuqebizo.bash | 16:42 |
| Mister_Magister | /usr/bin/droid/droid-load-firmware.sh is failing https://paste.opensuse.org/pastes/4ec89848b17a | 16:42 |
| Mister_Magister | but everything mounted and droid-hal started so that's nice | 16:43 |
| mal | Mister_Magister: check journal for mount errors | 16:43 |
| Mister_Magister | i checked systemd and not a single service failed | 16:43 |
| mal | is that file anywhere on the device? | 16:44 |
| Mister_Magister | oh sorry my bad | 16:44 |
| Mister_Magister | vendor_bt-firmware etc failed | 16:44 |
| mal | probably if vendor mount failed then it tried to mount that on top or dependency issue because of failed vendor mount | 16:44 |
| mal | reboot and see if fixing the vendor mount service helps with that | 16:45 |
| Mister_Magister | ah it's trying to mount /dev/block/bootdevice/by-name/bluetooth | 16:45 |
| Mister_Magister | which doesn't exist | 16:45 |
| Mister_Magister | fixup-mountpoints didn't work out | 16:45 |
| mal | Mister_Magister: ah, fixup issue | 16:45 |
| Mister_Magister | yeap | 16:45 |
| mal | Mister_Magister: often fstab has bluetooth but you probably did bluetooth_a in fixup | 16:46 |
| Mister_Magister | you got it :P | 16:46 |
| mal | check all partitions in fstab | 16:46 |
| mal | probably modem and other stuff could have same issue | 16:46 |
| T42 | <elros34> @the_hiktor ./out/target/product/surya/system/apex/com.android.runtime/ looks sane for me => has all libs both 32 and 64bit | 16:47 |
| Mister_Magister | should i remove all _b and remove _a? | 16:47 |
| mal | only remove _a from the ones in fstab | 16:48 |
| mal | check https://github.com/mer-hybris/hybris-boot/blob/master/fixup-mountpoints#L17 for example | 16:48 |
| Mister_Magister | oke | 16:48 |
| Mister_Magister | tho what's preventing me from just removnig all _a | 16:48 |
| mal | well nothing | 16:48 |
| Mister_Magister | :) | 16:48 |
| Mister_Magister | it's faster than doing it selectively | 16:48 |
| mal | unless some partition in fstab or elsewhere has it hardcoded | 16:49 |
| mal | probably grepping mount services on device tells that | 16:49 |
| mal | I think it mentions the original fstab/rc file entry in the comments | 16:49 |
| Mister_Magister | not a single one has _a in fstab | 16:50 |
| mal | ok | 16:50 |
| mal | also check the metadata in fixup that it's correct, often it uses different path | 16:51 |
| mal | or check it on device to see if the mount service looks correct | 16:51 |
| Mister_Magister | block/bootdevice/by-name/bluetooth sdd15 looks correct not sure what to check | 16:51 |
| mal | does metadata.mount have correct path? | 16:52 |
| Mister_Magister | good catch | 16:53 |
| Mister_Magister | What=/dev/block/by-name/metadata | 16:53 |
| mal | most devices now have it like that | 16:53 |
| mal | like in the pdx235 example I linked | 16:53 |
| Mister_Magister | i grepped for all What= and that was the only one | 16:54 |
| Mister_Magister | fixed it in fixup mountpoints | 16:54 |
| Mister_Magister | i'll have to add vendor.mount to sparse later | 16:54 |
| Mister_Magister | to correct the fs type | 16:54 |
| Mister_Magister | or i could edit it in device | 16:55 |
| mal | I nowadays do packaged droid-system on my devices | 16:55 |
| mal | so I don't have to mount vendor and friends | 16:55 |
| Mister_Magister | heh | 16:55 |
| Mister_Magister | hmm in fstab it has vendor /vendor ext4 | 16:56 |
| Mister_Magister | idk where it's got erofs from | 16:56 |
| mal | not sure, unless it found some other fstab also | 16:56 |
| mal | the rpm contains system, vendor, system_ext and odm, and any other needed things from super partition | 16:56 |
| Mister_Magister | there's on odm on this phone | 16:56 |
| mal | stripped down version of those partitions, removing all apk stuff etc | 16:57 |
| Mister_Magister | hmm vendor.mount doesn't have the comment where it's from | 16:57 |
| Mister_Magister | oh | 16:57 |
| Mister_Magister | i have it in sparse | 16:57 |
| mal | oops | 16:57 |
| Mister_Magister | yeap | 16:57 |
| mal | I was about to ask you to check that | 16:58 |
| Mister_Magister | whats fascinating i copied it from another of my ports and it works there | 16:58 |
| mal | maybe some device had it as erofs | 16:58 |
| Mister_Magister | apparently | 16:58 |
| Mister_Magister | yeah it's mounted as erofs | 16:58 |
| Mister_Magister | magic | 16:58 |
| mal | maybe after those mounts it will work a bit better, there are some a15 and 16 changes still to be done | 16:59 |
| mal | suspendservice mostly | 16:59 |
| Mister_Magister | ohno there's no suspend? but it had such a great battery life on android | 17:00 |
| Mister_Magister | i'll rebuild and repackage everything now so brb in one hour | 17:00 |
| Mister_Magister | maybe it will get to gui this time | 17:00 |
| mal | no, the suspend service just makes sure android side processes can prevent suspend if needed | 17:00 |
| Mister_Magister | oh | 17:01 |
| Mister_Magister | i wonder if it will have any good battery life on sfos | 17:01 |
| mal | in the past they used wakelocks but now they do it a bit differently so we have to patch suspendservice | 17:01 |
| Mister_Magister | also it's also 120Hz so the same 2x animation speed bug :))))))))))) | 17:01 |
| mal | I'll try to get that fixes for 5.1 | 17:02 |
| mal | *fixed | 17:02 |
| Mister_Magister | <3 thank you so much | 17:02 |
| Mister_Magister | i bought the device spontaniously but i started to like it, it's the most compact device (from sane vendor) in recent years so it's kinda decent to hold | 17:03 |
| Mister_Magister | any progress with camera2 api? | 17:04 |
| Mister_Magister | the cameras are being really held back by the denoising and sharpening | 17:04 |
| mal | it's closer, worked on some devices already | 17:04 |
| Mister_Magister | with video recording? | 17:04 |
| mal | yes | 17:05 |
| Mister_Magister | whoa | 17:05 |
| Mister_Magister | then when i'll be done with this port i'll bother you about doing the camera2 api on this phone too, is that okay? | 17:05 |
| mal | if you want to build suspendservice run "make android.system.suspend-service" | 17:05 |
| Mister_Magister | oki doki | 17:05 |
| mal | if you have latest hybris-patches the patch should be there | 17:06 |
| Mister_Magister | it's fresh port so yeah | 17:06 |
| mal | I pushed the change about a week ago | 17:06 |
| Mister_Magister | i bought the phone on tuesday | 17:07 |
| mal | or did I push it, I'll check | 17:08 |
| mal | looks like it's missing from hybris-22.2 branch | 17:08 |
| Mister_Magister | :) | 17:09 |
| mal | sorry | 17:09 |
| Mister_Magister | nah don't apologize | 17:09 |
| mal | too many things to do | 17:09 |
| Mister_Magister | yeah i get that | 17:10 |
| Mister_Magister | mal: if latest version is .77 and minor versions are not used, as long as i have _5.0 repo, people should be able to update to .77 right? | 17:11 |
| Mister_Magister | hmm .67 uses minor version still | 17:12 |
| mal | unless they update from < 5.0 | 17:12 |
| Mister_Magister | oh no nvm it is _5.0 | 17:13 |
| mal | if people update from older releases then they need to first update to one with the full version repo | 17:14 |
| mal | I mean from pre-5.0 releases but probably everyone has some 5.0 already | 17:15 |
| Mister_Magister | ye | 17:16 |
| T42 | <the_hiktor> @elros34 i understand but couldn't figure out where to do it :/ | 17:16 |
| Mister_Magister | but if they are the update will fail because repo doesn't exist | 17:16 |
| Mister_Magister | so we good | 17:16 |
| T42 | <elros34> @the_hiktor not sure if this is correct but you may try removing '.release' from here https://github.com/mer-hybris/droid-hal-device/blob/master/droid-hal-device.inc#L745 | 17:18 |
| T42 | <the_hiktor> lemme try | 17:18 |
| Mister_Magister | mal: my users have comfort of using my updater so if i simply change version on a website they will now be able to update to .77 with me doing literally nothing else :D | 17:27 |
| T42 | <the_hiktor> @elros34 nope didn't help 🥲 | 17:33 |
| Mister_Magister | me hoarding all the workers again because i updated droid-configs on 6 ports | 17:37 |
| Mister_Magister | wait i lost one, | 17:37 |
| T42 | <adampigg> is .77 public yet? | 17:42 |
| mal | https://forum.sailfishos.org/t/release-notes-tampella-5-0-0-77/28520 | 17:43 |
| mal | yes it is | 17:43 |
| mal | the list of 5.0 releases is getting quite long | 17:43 |
| T42 | <the_hiktor> @elro | 17:44 |
| T42 | <adampigg> cool i missed that, thought it was still ea | 17:44 |
| T42 | <adampigg> i guess there is no change from 76 to 77 for ports, just eas? | 17:44 |
| mal | yeah | 17:44 |
| T42 | <the_hiktor> @elros34 wait it did something https://hastebin.com/share/asohuxujod.bash | 17:45 |
| mal | it seems to detect kernel version incorrectly | 17:46 |
| Mister_Magister | yeah back in the day jolla was making one major release every 3 months | 17:47 |
| Mister_Magister | we've been at 5.0 for like a year now? | 17:47 |
| T42 | <adampigg> thats fine, im not accepting a 5.1 until sb2 changes are merged 😂 | 17:47 |
| Mister_Magister | less work for me | 17:48 |
| Mister_Magister | updating 7 ports every 3 months would get tedious fast | 17:48 |
| mal | @the_hiktor check the paths in line 440 of that log and see which one has different version, maybe something was not rebuilt | 17:48 |
| * Mister_Magister this port is gonna be my 7th | 17:48 | |
| mal | have you run only make hybris-boot at some point? | 17:48 |
| mal | next release after 5.1 will not be very long after that | 17:49 |
| Mister_Magister | ohnoo | 17:49 |
| Mister_Magister | more work for me | 17:50 |
| Mister_Magister | at least now i got op6, pipa and g2 users to test for me | 17:50 |
| T42 | <the_hiktor> mal, checked all the mentioned paths this looks diff | 18:08 |
| T42 | <the_hiktor> /mnt/Imp/hadk/out/target/product/surya/obj/KERNEL_OBJ/include/config/kernel.release | 18:08 |
| T42 | <the_hiktor> 4.14.246-lineageos-gfc4ead1b9162-dirty | 18:08 |
| T42 | <the_hiktor> | 18:08 |
| T42 | <the_hiktor> while the others | 18:08 |
| T42 | <the_hiktor> 4.14.246-lineageos-gde658ac8a055-dirty | 18:09 |
| mal | @the_hiktor did you run just make hybris-boot at some point? | 18:36 |
| mal | instead of running make hybris-hal | 18:36 |
| Mister_Magister | mal: flashed new image, i'm not as lucky as to boot straight to gui sadly so debugging shall commence | 18:41 |
| Mister_Magister | everything is mounted this time | 18:41 |
| Mister_Magister | huh, there's no defaultuser user… is that something being done on startup? | 18:43 |
| T42 | <the_hiktor> mal, yes I did after finishing with defconfig yesterday | 18:43 |
| Mister_Magister | /usr/bin/droid/wait_for_keymint: line 3: binder-list: not found hm | 18:46 |
| Mister_Magister | if i'm not mistaken that'd be https://github.com/VerdandiTeam/droid-config-pipa/blob/master/rpm/droid-config-pipa.spec#L25 | 18:46 |
| Mister_Magister | there's A LOT of audit logs | 18:48 |
| Mister_Magister | that normal? | 18:48 |
| Mister_Magister | https://paste.opensuse.org/pastes/dd24c95d8a8b | 18:49 |
| Mister_Magister | a freaking lot of avc: denied | 18:49 |
| Mister_Magister | can I add audit=0? | 18:51 |
| Mister_Magister | in the a15 we no longer disable selinux hence the question | 18:52 |
| Mister_Magister | oke so keymint is waiting properly now | 18:59 |
| Mister_Magister | qseecomd and keystore 2 seem to be bootlooping | 18:59 |
| Mister_Magister | not sure if it should be denying all of that | 19:00 |
| Mister_Magister | there's still the Could not find 'android.hardware.media.c2@1.0::IComponentStore/default' for ctl.interface_start | 19:00 |
| Mister_Magister | and logcat is unusable with 10k messages per second | 19:01 |
| Mister_Magister | sscrpcd is mostly spamming | 19:03 |
| Mister_Magister | there's also some denied stuff for sscrpcd | 19:05 |
| Mister_Magister | there isn't much mention in logs about sscrpcd :/ | 19:07 |
| Mister_Magister | ah okay so because selinux is in permissive mode the denied can be ignored | 19:09 |
| mal | @the_hiktor try building make hybris-hal and hopefully that helps with the mismatch | 19:10 |
| mal | Mister_Magister: audio=0 is ok, and recommended | 19:11 |
| Mister_Magister | oki thanks | 19:11 |
| Mister_Magister | ok missing services are keymaster2, odsign, vendor.qseecomd, keymint | 19:12 |
| Mister_Magister | mal: any idea how to debug it further? | 19:12 |
| Mister_Magister | i'm running thin on ideas | 19:12 |
| mal | Mister_Magister: is that pipa or some other device? | 19:12 |
| Mister_Magister | the miami i mentioned recently, motorola edge 30 neo i bought on tuesday | 19:13 |
| mal | ah, sorry | 19:13 |
| Mister_Magister | fresh port on android 15, pipa was android 14 so i look at it as reference | 19:13 |
| Mister_Magister | it could cause confusion | 19:13 |
| mal | check what you have in /etc/gbinder.conf | 19:13 |
| Mister_Magister | ApiLevel = 35 | 19:13 |
| mal | ok, should be ok | 19:13 |
| mal | so is binder-list still missing? | 19:14 |
| Mister_Magister | no, i added the requires above, same way pipa had it, and now we have binder list and as i mentioned "oke so keymint is waiting properly now" so that problem is resolved | 19:14 |
| Mister_Magister | not that keymint is running | 19:14 |
| Mister_Magister | just the script is waiting correctly | 19:14 |
| mal | Apr 08 11:12:23 localhost bluebinder_wait.sh[12234]: linker_tls.cpp:93: unregister_tls_module CHECK 'mod.static_offset == SIZE_MAX' failedAborted (core dumped) | 19:15 |
| mal | is there keymaster running? | 19:15 |
| Mister_Magister | well it's bluebinder so we can ignore bluetooth for now no? for reference the state is no gui or defaultuser | 19:15 |
| Mister_Magister | keymaster2 seems to be bootlooping | 19:15 |
| mal | I mean that tls issue will cause some problems | 19:16 |
| Mister_Magister | gotcha | 19:16 |
| Mister_Magister | hmm i don't think keymaster is bootlooping anymore | 19:16 |
| Mister_Magister | 1892 system /vendor/bin/hw/android.hardware.keymaster@4.1-service-qti | 19:16 |
| Mister_Magister | seems to be running (low pid so no restarts) | 19:16 |
| mal | maybe you need use wait_for_keymaster then | 19:17 |
| mal | like in a14 | 19:17 |
| Mister_Magister | alrighty then | 19:17 |
| Mister_Magister | i'll steal it from pipa | 19:17 |
| mal | and add keymint to delete files list | 19:17 |
| Mister_Magister | pipa doesn't have it | 19:17 |
| mal | I mean for the new device | 19:17 |
| Mister_Magister | ye i know i just thought i can steal it from pipa | 19:18 |
| Mister_Magister | but it doesn't have it | 19:18 |
| Mister_Magister | do you have example for the wait for keymaster service and the keymind delete files? | 19:18 |
| mal | other you will have both | 19:18 |
| mal | the a14 submodule has keymaster | 19:19 |
| Mister_Magister | ohh | 19:19 |
| mal | I thought a15 always has keymint but seems not | 19:19 |
| Mister_Magister | https://github.com/mer-hybris/droid-hal-configs/blob/5a5f6ce707889ef1bebc345425a0462e2218e0e9/sparse-14/usr/lib/systemd/system/wait_for_keymaster.service found it | 19:19 |
| Mister_Magister | and how do i add keymint to delete files? | 19:20 |
| Mister_Magister | oh the wait_for_keymaster file is present just service is missing | 19:20 |
| Mister_Magister | I'll change it because my keymaster is 4.1 and the file has 4.0 | 19:21 |
| Mister_Magister | i'll mask wait_for_keymint for now | 19:21 |
| mal | for delete files add to droid-configs root dir file delete_file_$DEVICE.list with the needed things you want to delete | 19:23 |
| mal | like here https://github.com/mlehtima/droid-config-fp5/blob/master/delete_file_fp5.list | 19:23 |
| Mister_Magister | ah thanks | 19:23 |
| Mister_Magister | hmm i enabled the service and masked keymint but that didn't stick after reboot :/ | 19:24 |
| Mister_Magister | oh | 19:24 |
| Mister_Magister | i masked keymaster xd | 19:24 |
| mal | oops | 19:24 |
| Mister_Magister | yeah :D | 19:24 |
| mal | probably good to add audit=0 already | 19:25 |
| mal | to get more reasonable logs | 19:25 |
| Mister_Magister | yeah the kernel is building | 19:25 |
| Mister_Magister | in the background | 19:25 |
| Mister_Magister | okay so wait_for_keymaster doesn't find keymaster, keymaster is running but binder-list -d /dev/hwbinder doesn't list it | 19:26 |
| mal | you could try pr-testing branch of libhybris from https://github.com/mer-hybris/libhybris to see if it helps with the getprop error | 19:26 |
| mal | wait | 19:26 |
| mal | check /dev/binder | 19:26 |
| mal | if it has keymaster or keymint | 19:27 |
| Mister_Magister | /dev/binder is literally empty | 19:27 |
| mal | hmm | 19:27 |
| Mister_Magister | binder-list -d /dev/binder | 19:27 |
| mal | that's not good | 19:27 |
| Mister_Magister | well yeah | 19:27 |
| mal | is servicemanager running? | 19:27 |
| Mister_Magister | it is | 19:28 |
| Mister_Magister | hwservicemanager and vndservicemanager also | 19:28 |
| mal | and you have latest libgbinder? | 19:28 |
| mal | probably not | 19:29 |
| mal | you need it | 19:29 |
| Mister_Magister | i just added the linked Requires and built image | 19:29 |
| Mister_Magister | so its the one from repos | 19:29 |
| mal | has to be 1.1.45 | 19:29 |
| mal | for a15+ | 19:29 |
| mal | forgot to mention that | 19:30 |
| Mister_Magister | it's 1.1.40 | 19:30 |
| Mister_Magister | so do i have to build it on obs? | 19:30 |
| mal | yes | 19:30 |
| Mister_Magister | gotcha | 19:30 |
| Mister_Magister | will do after eating | 19:30 |
| mal | wait | 19:30 |
| mal | I'll bump it hw common | 19:30 |
| mal | so just need to rebuild after that | 19:31 |
| Mister_Magister | <3 | 19:32 |
| mal | so maybe you actually have keymint but it was just not found because of too old gbinder | 19:34 |
| Mister_Magister | but it's not running | 19:37 |
| Mister_Magister | and it's not in vendor/etc/init nor system/etc/init | 19:38 |
| Mister_Magister | and I didn't see mention of it failing either | 19:38 |
| Mister_Magister | so i'd say unlikely | 19:39 |
| Mister_Magister | tho there are some so's | 19:39 |
| Mister_Magister | mal: may i build image now? | 19:44 |
| Mister_Magister | i'll take that as a yes (there's 1.1.45 in repos now) | 19:49 |
| mal | yes | 19:57 |
| T42 | <the_hiktor> mal, tried building hybris-hal it didnt help 😐 | 20:03 |
| Mister_Magister | mal: okay, not much more https://paste.opensuse.org/pastes/2eec9a1d020a | 20:20 |
| Mister_Magister | no keymint or keymaster | 20:20 |
| Mister_Magister | okay keystore2 seems to bootloop | 20:22 |
| Mister_Magister | https://paste.opensuse.org/pastes/4f5283f2ae51 | 20:22 |
| Mister_Magister | it says starting and then exited | 20:22 |
| Mister_Magister | when ran by hand i get CANNOT LINK EXECUTABLE "/system/bin/keystore2": library "libandroidicu.so" not found: needed by /system/lib64/libsqlite.so in namespace (default) | 20:23 |
| Mister_Magister | wait, libandroidicu i heard that name before | 20:23 |
| Mister_Magister | oh yee hadk-faq has info about this | 20:24 |
| Mister_Magister | libandroidicu.so are only in /apex/com.android.i18n/lib64/libandroidicu.so | 20:26 |
| mal | do you have odm? | 20:26 |
| mal | but odd that some android side process would have issues with that | 20:26 |
| Mister_Magister | i don't | 20:26 |
| Mister_Magister | I even checked when I was on android there's no odm mounted to speak of | 20:27 |
| mal | is /linkerconfig ok? | 20:27 |
| Mister_Magister | i mean it is mounted | 20:27 |
| Mister_Magister | idk if its okay | 20:27 |
| Mister_Magister | oh the file is empty | 20:27 |
| mal | but is the ld.config.txt in it ok | 20:27 |
| Mister_Magister | it's emtpy | 20:27 |
| mal | yeah, so no wonder things fail | 20:28 |
| Mister_Magister | what do we do captain | 20:28 |
| mal | question is why | 20:28 |
| Mister_Magister | idk what creates or writes to it | 20:28 |
| mal | in that android base /usr/libexec/droid-hybris/system/bin/linkerconfig | 20:28 |
| mal | check logs for linkerconfig | 20:29 |
| Mister_Magister | /usr/libexec/droid-hybris/system/bin/linkerconfig doesn't exist | 20:29 |
| mal | hmm | 20:29 |
| Mister_Magister | there's /usr/libexec/droid-hybris/system/bin/linker | 20:29 |
| mal | which rpm/dhd submodule revision do you have? | 20:30 |
| Mister_Magister | probably master | 20:30 |
| Mister_Magister | i made it very recently | 20:30 |
| mal | check the exact revision just in case | 20:30 |
| Mister_Magister | i'm already on it | 20:30 |
| Mister_Magister | 21cc677be30b4adb152843d51265ae4bb0641a39 | 20:30 |
| Mister_Magister | Copy linkerconfig always if it exists | 20:30 |
| Mister_Magister | by you :) | 20:30 |
| Mister_Magister | should I make linkerconfig and repackage? | 20:30 |
| mal | check out for linkerconfig | 20:31 |
| mal | it should have been built, maybe it's in wrong place | 20:32 |
| Mister_Magister | i'm already running find out/ -name linkerconfig | 20:32 |
| Mister_Magister | out/target/product/miami/root/linkerconfig | 20:32 |
| Mister_Magister | https://paste.opensuse.org/pastes/b03d67a5726c here's all of them | 20:32 |
| mal | hmm | 20:33 |
| Mister_Magister | wrong path | 20:33 |
| mal | but there is not proper path | 20:33 |
| Mister_Magister | there's /symbols/apex/com.android.runtime/bin/linkerconfig but not /apex/com.android.runtime/bin/linkerconfig | 20:33 |
| mal | yeah, but symbols is not a proper place | 20:33 |
| Mister_Magister | is there like some patch missing or something? | 20:34 |
| mal | try make linkerconfig | 20:34 |
| Mister_Magister | trying | 20:34 |
| mal | @the_hiktor that is strange, one option is the remove out/target/product/ and rebuild hybris-hal but that of course will take some time | 20:36 |
| Mister_Magister | mal: nothing changed, build said "Install: out/host/linux-x86/bin/linkerconfig" but out/target/product/miami/ did not change | 20:38 |
| mal | hmm | 20:39 |
| Mister_Magister | there's no out/target/product/miami/apex whatsoever | 20:39 |
| Mister_Magister | not even empty directory | 20:39 |
| mal | try make linkerconfig.com.android.runtime | 20:41 |
| mal | no idea if that can work but just guessing | 20:41 |
| Mister_Magister | sorry for stupid question but can't we just change path? | 20:42 |
| Mister_Magister | ninja: no work to do. | 20:42 |
| mal | yes, but that would be ugly | 20:43 |
| mal | out/target/product/miami/obj/EXECUTABLES/linkerconfig.com.android.runtime_intermediates/linkerconfig is the correct file in your case | 20:44 |
| Mister_Magister | not root/linkerconfig? | 20:45 |
| mal | that is target folder i.e. /linkerconfig on device | 20:46 |
| mal | afaik | 20:46 |
| Mister_Magister | but like out/target/product/miami/root/linkerconfig exists | 20:47 |
| mal | but it's a folder | 20:47 |
| mal | not a program | 20:47 |
| Mister_Magister | oh… right | 20:47 |
| Mister_Magister | also /apex/com.android.runtime/bin/linkerconfig exists on device | 20:48 |
| Mister_Magister | also out/target/product/miami/symbols/apex/com.android.runtime/bin/linkerconfig is also executable | 20:48 |
| Mister_Magister | so why not use the symbols subdirectory/ | 20:48 |
| mal | I said out/target/product/miami/obj/EXECUTABLES/linkerconfig.com.android.runtime_intermediates/linkerconfig is the correct one | 20:52 |
| mal | the symbols one is huge | 20:52 |
| Mister_Magister | oh okay | 20:52 |
| mal | but that doesn't solve the root cause | 20:52 |
| Mister_Magister | just asking stupid questions | 20:52 |
| Mister_Magister | can't ssh so can't get the file onto device without playing with telnet so lets figure out why is it in wrong path | 20:54 |
| mal | you can copy files easily via telnet | 20:57 |
| mal | well somewhat easily | 20:58 |
| Mister_Magister | i know i know | 20:58 |
| Mister_Magister | but it's annoying | 20:58 |
| Mister_Magister | and we gotta figure it out one way or another | 20:58 |
| mal | yeah, netcat is not best tool to use | 20:58 |
| mal | I think for now just hack the path in droid-hal-device.inc | 20:58 |
| Mister_Magister | okay i'll copy the file… | 20:58 |
| mal | just fix the path in rpm/dhd | 20:59 |
| mal | of course copying it in out is ok also | 20:59 |
| Mister_Magister | oh | 21:00 |
| Mister_Magister | yeah but then i'd have to rebuild image and all, i'll just copy the executable to the device for now | 21:00 |
| Mister_Magister | okay /linkerconfig is full of files now | 21:03 |
| Mister_Magister | and ld | 21:03 |
| Mister_Magister | WE GOT GUI | 21:03 |
| Mister_Magister | let's check step two of setupwizard | 21:04 |
| Mister_Magister | funnily droid-late-start.service failed | 21:04 |
| Mister_Magister | init-done also isn't done | 21:04 |
| Mister_Magister | step two started | 21:04 |
| Mister_Magister | no wifi, ofono seems to be working | 21:05 |
| mal | did you reboot after adding the linkerconfig? | 21:05 |
| Mister_Magister | yes | 21:05 |
| mal | I assume you didn't yet try the pr-testing libhybris | 21:06 |
| Mister_Magister | what's that | 21:06 |
| mal | I think some issues are because of getprop and setprop failing due to the crash | 21:06 |
| mal | 22:26 < mal> you could try pr-testing branch of libhybris from https://github.com/mer-hybris/libhybris to see if it helps with the getprop erro | 21:06 |
| Mister_Magister | sorry i never saw that message | 21:07 |
| Mister_Magister | it's not in telegram chat either | 21:07 |
| Mister_Magister | so maybe error on your end | 21:07 |
| mal | nope https://irclogs.sailfishos.org/logs/%23sailfishos-porters/2026/%23sailfishos-porters.2026-04-10.log.html#t2026-04-10T19:26:40 | 21:08 |
| Mister_Magister | ayy it's so smooth | 21:08 |
| Mister_Magister | ahh different timezone | 21:08 |
| mal | yeah | 21:08 |
| mal | that is a wip branch for the jolla phone | 21:09 |
| Mister_Magister | yeah sorry your messages after that one distracted me :D | 21:09 |
| Mister_Magister | bluetooth works lol | 21:09 |
| mal | that earlier mention in log was indeed a bit confusing as I was talking about two different issues | 21:12 |
| Mister_Magister | I used bluetooth to transfer screenshot and it worked :D | 21:12 |
| Mister_Magister | sorry i'm currently too excited with the device i'll go back to that soon | 21:12 |
| Mister_Magister | wifi doesn't work probably due to missing module or something | 21:12 |
| mal | or missing echo to correct device node | 21:13 |
| Mister_Magister | ye | 21:13 |
| mal | maybe something like this https://github.com/mlehtima/droid-config-fp5/blob/master/sparse/usr/lib/systemd/system/wifisetup.service | 21:13 |
| mal | depends on device if it's like that or if some module is missing or what | 21:13 |
| Mister_Magister | ye i know i know | 21:14 |
| Mister_Magister | pipa needed echo | 21:14 |
| mal | most qcom devices need it now | 21:14 |
| mal | mediatek needs something different for example | 21:15 |
| Mister_Magister | I know you probably don't care but here's screenshot :3 https://pbs.twimg.com/media/HFksp7JXYAEgp9J?format=jpg&name=4096x4096 | 21:15 |
| Mister_Magister | it's so smoooth | 21:15 |
| Mister_Magister | even sim card works flawlessly (no 5G tho | 21:20 |
| mal | have you enabled 5g in csd config | 21:20 |
| mal | or was it needed | 21:21 |
| Mister_Magister | idk | 21:21 |
| mal | https://github.com/mer-hybris/droid-config-sony-murray/blob/master/sparse/usr/share/csd/settings.d/10-hw-settings.ini#L10 | 21:21 |
| mal | not sure if that affect anything else that csd tests | 21:22 |
| mal | but while at it check that the config is correct in your repo for all features | 21:22 |
| Mister_Magister | i was touching it so it should be somewhat correct | 21:23 |
| Mister_Magister | Cellular5G is =1 | 21:24 |
| Mister_Magister | maybe i don't have ofono binder plugin idk | 21:24 |
| Mister_Magister | anyway | 21:24 |
| Mister_Magister | libhybris | 21:24 |
| Mister_Magister | let me copy the linkerconfig in out | 21:24 |
| Mister_Magister | copied | 21:25 |
| mal | ofono-binder-plugin should be latest in 5.0 | 21:25 |
| mal | if you built at least .73 | 21:25 |
| Mister_Magister | echo ON > /dev/wlan did NOT enable wifi | 21:25 |
| mal | is there /dev/wlan? | 21:26 |
| mal | check dmesg | 21:26 |
| T42 | <the_hiktor> mal, https://hastebin.com/share/udusovamil.bash | 21:27 |
| Mister_Magister | cat /dev/wlan says ON so i don't think so :) | 21:27 |
| Mister_Magister | dmesg also didn't say a word | 21:27 |
| Mister_Magister | there's wlan.ko which i modprobed but that didn't help | 21:28 |
| Mister_Magister | there's a lot of dmesg spam so hard to tell anything from it | 21:29 |
| Mister_Magister | ok i'll try libhybris now | 21:32 |
| Mister_Magister | mal: do i have to rebuild hal or just obs change? | 21:32 |
| mal | obs is enough | 21:32 |
| mal | @the_hiktor I thought you already managed to build it once? | 21:33 |
| Mister_Magister | https://build.sailfishos.org/package/rdiff/nemo:testing:hw:motorola:miami/libhybris?linkrev=base&rev=3 | 21:33 |
| Mister_Magister | in case you want to validate | 21:33 |
| T42 | <the_hiktor> mal, I'm stuck on building droid-hal 🙃 | 21:35 |
| Mister_Magister | mal: hmm any chances of getting wifi to work? i could then just zypper ref zypper dup instead of rebuilding | 21:36 |
| T42 | <the_hiktor> mal, Idk what else I'm supposed to do or say🥲 | 21:36 |
| mal | Mister_Magister: you could always share network over usb until you get wlan working | 21:38 |
| Mister_Magister | yeah but that's even more annoying and it didn't really work for me | 21:38 |
| mal | @the_hiktor the error is strange, I don't know where it comes from | 21:38 |
| mal | @the_hiktor have you done any changes to file rpm/dhd/droid-hal-device.inc? | 21:39 |
| mal | Mister_Magister: no wlan interface? | 21:40 |
| Mister_Magister | nope | 21:40 |
| mal | so the echo did nothing? are you really sure not showed in dmesg for example | 21:41 |
| Mister_Magister | ohhhhh | 21:41 |
| Mister_Magister | mal i got it | 21:41 |
| mal | ? | 21:41 |
| Mister_Magister | i FIRST did modprobe wlan and then /dev/wlan appeared and echo ON worked! | 21:41 |
| mal | ah | 21:41 |
| Mister_Magister | but this time it didn't feel like booting into gui | 21:41 |
| Mister_Magister | oh it booted | 21:41 |
| T42 | <the_hiktor> mal, That's where I've been missing around since yesterday 🙃 | 21:41 |
| Mister_Magister | wifi works! | 21:41 |
| Mister_Magister | got no vibration and no sound, this is gonna be fun (not) | 21:42 |
| mal | go to that rpm/dhd folder and run "git diff" and paste it to some service | 21:42 |
| mal | Mister_Magister: check evdev_trace -i to see if it has memless, if not then check permissions of the sysfs vibra led paths | 21:43 |
| Mister_Magister | how do i know if it has memless? | 21:43 |
| Mister_Magister | https://paste.opensuse.org/pastes/6803c2fa8086 | 21:43 |
| mal | Mister_Magister: it doesn't based on that, so check the sysfs permissions | 21:44 |
| Mister_Magister | do you have path on hand perhaps? im not sure what to look for | 21:44 |
| Mister_Magister | updated libhybris, rebooting | 21:44 |
| mal | Mister_Magister: something like /sys/class/leds/vibrator | 21:44 |
| mal | some devices have different last part | 21:45 |
| T42 | <the_hiktor> mal, here https://hastebin.com/share/nutikiqona.bash | 21:45 |
| Mister_Magister | oki will check | 21:45 |
| mal | @the_hiktor the line 39 in that paste is wrong, undo the changes shown in lines 37-39 | 21:46 |
| mal | also the lines 47-> look quite wrong | 21:47 |
| Mister_Magister | and i'm ssh over wifi phew | 21:47 |
| mal | Mister_Magister: for the wlan module maybe you need modules.load.d file | 21:47 |
| Mister_Magister | mal: another stupid question for you, when making wifisetup service, should i add modprobe as ExecPre or should i play with /etc/modules.laod | 21:48 |
| mal | Mister_Magister: like this https://github.com/mlehtima/droid-config-fp5/blob/master/sparse/etc/modules-load.d/fp5.conf | 21:48 |
| Mister_Magister | literally same brain | 21:48 |
| Mister_Magister | great minds think alike | 21:48 |
| mal | as usual if in doubt check my devices :) | 21:48 |
| Mister_Magister | :D | 21:48 |
| Mister_Magister | i'll check if the echo is needed at all | 21:49 |
| Mister_Magister | okay with the new libhybris init-done is done only policies-setup.service failed | 21:50 |
| Mister_Magister | keymint is still waiting | 21:50 |
| Mister_Magister | keystore is there but not keymint https://paste.opensuse.org/pastes/c0032a568d8f | 21:51 |
| Mister_Magister | so i think it really is missing and i should yeet wait for keymint for sake of wait for keymaster | 21:51 |
| T42 | <the_hiktor> mal, I didn't touch those 37 to 39 but I did at 47 | 21:51 |
| mal | @the_hiktor maybe accident then, but make sure lines shows in 37 is there and not the changes in 38-39 | 21:53 |
| Mister_Magister | ok echo > dev/wlan is needed | 21:53 |
| Mister_Magister | Who was I to ping about store access? | 21:56 |
| Mister_Magister | can't quite find that info, the new hadk says to just ping the chat xd | 21:57 |
| Mister_Magister | wait_for_keymaster.service works fine | 21:58 |
| Mister_Magister | and with the wifi service it also works fine | 21:58 |
| Mister_Magister | hmm there's not a vibrator file in entirety of /sys | 21:59 |
| Mister_Magister | there's nothing in leds folder | 22:00 |
| Mister_Magister | i wonder if it's another module | 22:01 |
| Mister_Magister | it is aw862x.ko | 22:02 |
| Mister_Magister | when i modprobed it, device vibrated xd | 22:02 |
| mal | Mister_Magister: check the modules.load things in various lib/modules folders (system_dlkm or whatever) and add the modules to modules-load.d | 22:04 |
| Mister_Magister | shall do | 22:05 |
| mal | Mister_Magister: maybe check out/ for .ko files and see those folders for load lists | 22:05 |
| mal | Mister_Magister: list of some possible places https://github.com/mer-hybris/droid-hal-device/blob/master/droid-hal-device.inc#L1058 | 22:06 |
| Mister_Magister | mal: i added the vibrator module to modules-load but it still doesn't vibrate, i chown'ed everything to be system:input and no luck either, any ideas? | 22:07 |
| mal | check evdev_trace -i again | 22:07 |
| Mister_Magister | it's exactly the same | 22:08 |
| Mister_Magister | i have the /sys/class/leds/vibrator/ and files in it | 22:08 |
| mal | what kind of files | 22:08 |
| mal | I wonder if you have one of those complex vibrators like fp5 does | 22:09 |
| Mister_Magister | https://paste.opensuse.org/pastes/5da48dbfc07e | 22:09 |
| Mister_Magister | uhh looks complex | 22:09 |
| mal | yeah | 22:09 |
| mal | maybe you need the binder vibra plugin | 22:09 |
| Mister_Magister | oh my | 22:09 |
| mal | https://github.com/mer-hybris/ngfd-plugin-droid-vibrator/tree/aidl | 22:10 |
| Mister_Magister | mal: i have out/blablabla/vendor/lib/modules/modules.load and it seems to contain bulk of the onboard modules. should i put that as modules-load file? | 22:11 |
| mal | yeah, does it have wlan and vibra drivers in there also | 22:11 |
| Mister_Magister | man i thought this device doesn't have LED, you'll not believe in what useless freaking place they put it | 22:12 |
| mal | ? | 22:12 |
| T42 | <the_hiktor> mal, 🥲 https://hastebin.com/share/biyakubivu.bash | 22:13 |
| Mister_Magister | mal: they put it on camera island | 22:13 |
| Mister_Magister | you will literally never see it | 22:13 |
| Mister_Magister | it's next to useless | 22:13 |
| mal | @the_hiktor you didn't fix it like I told | 22:13 |
| Mister_Magister | mal: it does have wlan and vibra drivers at the very least | 22:14 |
| Mister_Magister | and the pointlesslyuseless led too | 22:14 |
| mal | @the_hiktor https://github.com/mer-hybris/droid-hal-device/blob/master/droid-hal-device.inc#L878 make sure that line and the surrounding lines are unmodified | 22:14 |
| T42 | <the_hiktor> mal, sry i forget to fix the remaining lines | 22:20 |
| T42 | <the_hiktor> anyway https://hastebin.com/share/zazavadave.bash | 22:20 |
| T42 | <bruuuh121> /theme_pixel_ratio | 22:21 |
| Mister_Magister | mal: now led module loads with the rest but it never comes up during boot i wonder if it's due to some race condition? | 22:21 |
| T42 | <bruuuh121> /theme_pixel_ratio | 22:21 |
| Mister_Magister | camera doesn't even launch gui before it freezes on black screen | 22:22 |
| mal | @the_hiktor something like this but with the paths in your error https://github.com/mlehtima/droid-hal-fp5/blob/master/droid-hal-fp5.spec#L29 | 22:22 |
| Mister_Magister | light sensor in csd also just freezes the app | 22:22 |
| Mister_Magister | same with proximity | 22:22 |
| Mister_Magister | i think all sensors are dead | 22:22 |
| mal | Mister_Magister: you need the aidl sensors | 22:22 |
| Mister_Magister | gotcha | 22:22 |
| Mister_Magister | …how do i do that? xd | 22:23 |
| mal | jb61406 branch of sensorfw | 22:23 |
| Mister_Magister | o7 | 22:23 |
| mal | you only need to build the hybris spec | 22:23 |
| Mister_Magister | wait so i need sensorfw-hybris package on obs or just sensorfw will do? | 22:25 |
| Mister_Magister | ahh i got my answer when it failed to build | 22:26 |
| mal | hybris package should be fine i think | 22:27 |
| Mister_Magister | https://build.sailfishos.org/package/show/nemo:testing:hw:motorola:miami/sensorfw-qt5-hybris | 22:27 |
| Mister_Magister | hopefully all good | 22:27 |
| Mister_Magister | i'll now package the wifi fixes and the modules-load before i forget | 22:27 |
| Mister_Magister | oh and i have to fix usb networking still | 22:27 |
| Mister_Magister | i still have usb-moded masked | 22:27 |
| mal | check the paths and udc | 22:28 |
| Mister_Magister | but i don't think i'll bother you anymore as it's late, i'll polish the port with what we've learned and leave it for today and start fresh tomorrow | 22:28 |
| Mister_Magister | i have multiple pending changes i need to remember | 22:28 |
| Mister_Magister | the aidl vibrator i also have to add | 22:29 |
| Mister_Magister | osco copypac nemo:testing:hw:fairphone:fp5/ngfd-plugin-binder-vibrator nemo:testing:hw:motorola:miami | 22:29 |
| Mister_Magister | Easy fix :D | 22:29 |
| Mister_Magister | also i'll play with the scaling of the ui that needs fine tuning | 22:30 |
| mal | you also need new enough droid-hal-version submodule and pattern changes | 22:30 |
| mal | and the define in droid-hal-version spec, check fp5 | 22:30 |
| Mister_Magister | it's newer than tuesday so probably newest | 22:30 |
| Mister_Magister | will check | 22:30 |
| Mister_Magister | oh mal the modules must've fixed audio! | 22:33 |
| Mister_Magister | ohh decoding video is baaad but that's probably easy fix | 22:34 |
| Mister_Magister | oh boi i'm so happy i don't have to mess with the audio | 22:35 |
| mal | well you might have issues with call audio, check that | 22:35 |
| Mister_Magister | yeah | 22:35 |
| mal | I remember some devices don't like to work well without the halium audio module | 22:35 |
| Mister_Magister | op6 or pipa needed halium module, and x2 has broken in-call mic | 22:36 |
| Mister_Magister | so i'm fully aware :) | 22:36 |
| Mister_Magister | I only hope it's easier helping me due to mye experience :D | 22:36 |
| Mister_Magister | mal: have choice for hybris-libsensorfw-qt5: hybris-libsensorfw-qt5-binder hybris-libsensorfw-qt5-hal, i assume binder? | 22:38 |
| Mister_Magister | or the -version changes fixes that | 22:38 |
| mal | wrong package name | 22:39 |
| Mister_Magister | what do you mean? | 22:39 |
| mal | it should be sensorfw-qt5-binder | 22:39 |
| mal | sorry | 22:39 |
| Mister_Magister | i mean https://build.sailfishos.org/project/monitor/nemo:testing:hw:motorola:miami?arch_aarch64=1&defaults=0&repo_sailfishos_5_0=1&unresolvable=1 | 22:39 |
| Mister_Magister | thats what it says | 22:40 |
| mal | https://build.sailfishos.org/package/show/nemo:testing:hw:motorola:miami/sensorfw-qt5-hybris | 22:41 |
| mal | that needs to be named sensorfw-qt5-binder | 22:41 |
| Mister_Magister | ah | 22:41 |
| mal | that is how is selects the spec to use | 22:41 |
| Mister_Magister | i know | 22:41 |
| Mister_Magister | but you said hybris so hybris | 22:41 |
| mal | I said sorry | 22:42 |
| Mister_Magister | :D don't be | 22:42 |
| mal | remembered incorrectly the spec name | 22:42 |
| Mister_Magister | yeah happens don't sweat it | 22:42 |
| Mister_Magister | man the speakers suck in these phones nowadays :D | 22:43 |
| Mister_Magister | about the video it was something something quirks something something codec supplied with should fix it? | 22:43 |
| mal | depends on how it fails | 22:44 |
| Mister_Magister | like green smudge holdup | 22:47 |
| Mister_Magister | https://i.imgur.com/DO6Ef1j.png | 22:47 |
| Mister_Magister | like this | 22:47 |
| mal | https://github.com/sailfishos/gst-droid/blob/acc60b71274e29426f1046b06d5c99c1b9b4d0f7/gst-libs/gst/droid/gstdroidcodec.h#L32 try those | 22:48 |
| mal | it could be the color conversion also | 22:48 |
| Mister_Magister | exactly what i mentioned xd | 22:48 |
| Mister_Magister | op6 has thoses | 22:48 |
| Mister_Magister | oh, it already has some quirks | 22:49 |
| Mister_Magister | literally copy pasted them form op6 and its fine now | 22:52 |
| Mister_Magister | also no mic in calls but we'll worry about that when we get volte and 5G because it changes things | 22:53 |
| mal | so use halium module, it usually fixes mic issue in calls | 22:53 |
| Mister_Magister | yeah thanks | 22:53 |
| mal | I really need to figure out why it's needed | 22:54 |
| Mister_Magister | :D | 22:54 |
| Mister_Magister | I have a lot of stuff to tie now and i have all the answers so once again, thank you from the bottom of my hear for the assistance and i shall do the stuff | 22:54 |
| mal | or course the idea of doing binder based pulseaudio module would be even better | 22:54 |
| mal | I suspect it wouldn't even be terribly big thing | 22:55 |
| Mister_Magister | sensors fixed :) | 22:57 |
| Mister_Magister | oh | 23:01 |
| Mister_Magister | camera works in csd | 23:01 |
| Mister_Magister | hold on a second | 23:01 |
| Mister_Magister | CAMERA WORKS! | 23:01 |
| Mister_Magister | the codec stuff must've fixed it | 23:01 |
| mal | or was it before the module fixes when you tried it? | 23:02 |
| Mister_Magister | no after module fixes | 23:02 |
| Mister_Magister | i mean i tried it after module fixes | 23:02 |
| Mister_Magister | and it wasn't working | 23:02 |
| Mister_Magister | man everything… just… works | 23:03 |
| Mister_Magister | where have the times when we spend half a year debugging gui gone… | 23:03 |
| Mister_Magister | good old 2015 | 23:03 |
| mal | well the issues with new bases were fixed already | 23:04 |
| Mister_Magister | yeah but i have the feeling that new bases are super easy to port while in 2015 it was taking WAY longer and was WAY more difficult | 23:05 |
| mal | sometimes yes | 23:05 |
| mal | sometimes no | 23:05 |
| Mister_Magister | maybe just my devices then | 23:05 |
| mal | for example this is needed on jolla phone https://github.com/mlehtima/libhybris-1/commit/a1b75cb37fe50198a9a0353f416413464a13184e | 23:05 |
| Mister_Magister | but like pipa, and now this, everything just works, you just have to load correct things or use proper service and voila | 23:05 |
| Mister_Magister | no debugging | 23:05 |
| mal | qcom is usually relatively easy | 23:05 |
| Mister_Magister | none whatsoever | 23:05 |
| Mister_Magister | all my devices have been qcom | 23:06 |
| Mister_Magister | (and mostly motorola) | 23:06 |
| Mister_Magister | hmm ui sounds are insanely loud and not respecting ringtone volume :( | 23:08 |
| mal | I think I had some issues like that on some device | 23:08 |
| Mister_Magister | relatable :D | 23:09 |
| Mister_Magister | i'll go freeze outside for a second to test gps | 23:09 |
| Mister_Magister | didn't even have to freeze it instantly saw sattelites | 23:10 |
| mal | for nfc you probably need newer nfc binder plugin | 23:10 |
| mal | you seem to have aidl nfc | 23:10 |
| mal | not sure if you have aidl or hidl gnss | 23:11 |
| Mister_Magister | mal: before you go is there like something in ofono config needed for 5G? it is using ofono-binder-plugin | 23:11 |
| Mister_Magister | I don't even have means of testing nfc so | 23:12 |
| Mister_Magister | is there like phone with 5g i can look at for reference? | 23:12 |
| mal | nothing should be needed | 23:13 |
| Mister_Magister | hm | 23:13 |
| Mister_Magister | well there's no 5G option in settings | 23:13 |
| Mister_Magister | only 4G | 23:13 |
| mal | hmm | 23:13 |
| mal | show the binder plugin config | 23:13 |
| mal | for ofono | 23:13 |
| Mister_Magister | i think it's using default one i didn't override it | 23:13 |
| Mister_Magister | https://paste.opensuse.org/pastes/60290f7c1b36 like this | 23:14 |
| Mister_Magister | and similar for dual sim | 23:14 |
| mal | wait, which interface do you use? | 23:15 |
| mal | https://github.com/mlehtima/droid-config-fp5/blob/master/sparse/etc/ofono/binder.d/fp5.conf | 23:15 |
| mal | you never remember to check my devices | 23:15 |
| Mister_Magister | :D | 23:15 |
| Mister_Magister | sorry :D | 23:15 |
| Mister_Magister | so i did restarted ofono and 5g option is there but it firmly sits on 4G | 23:17 |
| mal | well often I have 4g on my device also | 23:17 |
| Mister_Magister | on android i tested it, it grabbed 5G instantly | 23:17 |
| mal | hmm | 23:17 |
| Mister_Magister | i will try reboot | 23:17 |
| Mister_Magister | alsoalso there's no battery charge thresholding options, is that normal? Some missing paths in sysfs? | 23:18 |
| mal | depends on the path you might need to configure it | 23:18 |
| Mister_Magister | could you give me a pointer as to the names of it as i have no clue | 23:18 |
| mal | https://github.com/mlehtima/droid-config-fp5/blob/master/sparse/etc/mce/70-battery-fp5.ini | 23:19 |
| Mister_Magister | thank you kindly | 23:19 |
| mal | it can be anything | 23:19 |
| Mister_Magister | well but i can now at least search for "charging" | 23:19 |
| Mister_Magister | reboot and it's 5G now | 23:19 |
| mal | just look the what is on the power_supply sysfs folders | 23:19 |
| Mister_Magister | and it dropped to 4G | 23:19 |
| Mister_Magister | ugh! | 23:19 |
| Mister_Magister | will do | 23:19 |
| mal | usually it only goes to 5g if there is need for mobile data | 23:19 |
| mal | at least I sometimes have it happen | 23:20 |
| Mister_Magister | oh it did | 23:20 |
| Mister_Magister | i got solid 40Mbps | 23:20 |
| Mister_Magister | nice | 23:20 |
| mal | I think that happen when it's non-standalone 5g | 23:20 |
| Mister_Magister | i probably forgot half the fixes i was supposed to apply at this point xd | 23:20 |
| Mister_Magister | also its sad that volte situation wasn't improved with 5G :( | 23:21 |
| mal | ? | 23:21 |
| Mister_Magister | ahh nothing | 23:21 |
| mal | there is the qcom plugin I use on fp4 and fp5 | 23:22 |
| Mister_Magister | yeyeye i know i know im using it on op6 before carrier blacklisted my imeis despite it working flawlessly for 2 days | 23:22 |
| mal | oh | 23:22 |
| Mister_Magister | yeah that situation | 23:22 |
| mal | very odd | 23:22 |
| Mister_Magister | that my 2021 android device also has broken volte so i bought 2022 device with hopefully working volte that i'm porting right now | 23:23 |
| Mister_Magister | and the only device in my collection that has working volte is chinese no-name car radio | 23:23 |
| mal | maybe some countries are doing stupid things with volte | 23:23 |
| Mister_Magister | let's just say i'm not happy | 23:23 |
| Mister_Magister | i'm in fact, quite upset | 23:23 |
| Mister_Magister | but you said yourself you don't like how volte is dependent on the carrier | 23:23 |
| mal | yeah | 23:23 |
| Mister_Magister | so that situation | 23:24 |
| mal | and of course the operator won't tell why they blacklisted it | 23:24 |
| mal | so we could fix the issue they might have noticed | 23:24 |
| Mister_Magister | yeah no they told me they're working with manufacturer bullshit | 23:24 |
| Mister_Magister | whats strangest is that even my zenfone 8 with android doesn't work | 23:24 |
| mal | wow | 23:25 |
| mal | so just shitty operator then | 23:25 |
| Mister_Magister | orange :) | 23:25 |
| Mister_Magister | anyway, the device will need some debugging about charging detection cause it's qi enabled | 23:25 |
| Mister_Magister | and here's the power_supply directory https://paste.opensuse.org/pastes/ea11cb13b711 | 23:26 |
| mal | did you find any path? | 23:26 |
| mal | but what do those contain? | 23:26 |
| Mister_Magister | ah sorry | 23:26 |
| Mister_Magister | battery at least has charge_control_limit_max | 23:26 |
| mal | just show them, I have checked those on several devices and have some idea what to look for | 23:27 |
| Mister_Magister | here you go https://paste.opensuse.org/pastes/487e4428212f | 23:27 |
| Mister_Magister | yeah was working on it | 23:27 |
| mal | what does /sys/class/power_supply/battery/input_current_limit contain? | 23:28 |
| Mister_Magister | 0 | 23:28 |
| mal | https://github.com/mer-hybris/droid-config-sony-zambezi/blob/master/sparse/etc/mce/70-battery-zambezi.ini | 23:28 |
| mal | :) | 23:28 |
| mal | check if echo 1 disables charging | 23:29 |
| Mister_Magister | lets' test | 23:29 |
| Mister_Magister | yeah | 23:29 |
| Mister_Magister | same brain | 23:29 |
| mal | you should be able to see the change in status in there | 23:30 |
| mal | or something | 23:30 |
| Mister_Magister | status is broken due to qi charging i think | 23:31 |
| Mister_Magister | it's constantly charging | 23:31 |
| mal | hmm | 23:31 |
| Mister_Magister | it charged 2 procent on charger and then i set value to 1 | 23:31 |
| Mister_Magister | let's see | 23:31 |
| Mister_Magister | nah it still charges | 23:31 |
| Mister_Magister | and QUITE FAST | 23:31 |
| Mister_Magister | i think setting it to 1 is bad idea | 23:32 |
| mal | how did you what does input_current_limit in usb do? | 23:32 |
| mal | check what it contains | 23:32 |
| Mister_Magister | instead of not charging i think it ignored current limiting and was charging insanly fast | 23:32 |
| mal | ok | 23:32 |
| Mister_Magister | like 1% per 2 seconds | 23:32 |
| Mister_Magister | ungodly fast at ~90% | 23:33 |
| mal | yeah, echo 0 | 23:33 |
| Mister_Magister | i did | 23:33 |
| mal | did it slow down? | 23:33 |
| Mister_Magister | yeap its sitting on 99% | 23:33 |
| mal | check the usb input_current_limit | 23:34 |
| Mister_Magister | -22 | 23:34 |
| mal | hmm | 23:34 |
| mal | how about dc one | 23:35 |
| Mister_Magister | anyway it's late if you need to go really don't feel like i'm holding you | 23:35 |
| mal | I'm watching a movie | 23:36 |
| Mister_Magister | dc is 1 | 23:36 |
| T42 | <the_hiktor> mal, done 😭 https://hastebin.com/share/amozokakem.bash | 23:36 |
| Mister_Magister | must be boring movie for you to type with me :D | 23:36 |
| mal | Mister_Magister: it's not | 23:36 |
| Mister_Magister | was a joke :P | 23:36 |
| mal | I'm used to multi-tasking | 23:36 |
| mal | @the_hiktor now build rest of the packages | 23:37 |
| mal | Mister_Magister: also the charger input_current_limit | 23:37 |
| mal | check all of those and see what those contain | 23:37 |
| Mister_Magister | charger does'nt have input_current_limit | 23:38 |
| Mister_Magister | i did cat */input_current_limit one has 0, some have 5, one has 100000, one 1 and one -22 | 23:38 |
| mal | your paste said charger has it | 23:39 |
| mal | line 118 | 23:39 |
| mal | which one is 1? | 23:39 |
| Mister_Magister | true | 23:39 |
| Mister_Magister | oh my mistake | 23:39 |
| Mister_Magister | charger is the 100000 one | 23:39 |
| mal | ok, not that then | 23:40 |
| Mister_Magister | dc has 1 | 23:40 |
| Mister_Magister | well we shall leave that for another day then | 23:42 |
| mal | let's check some other paths | 23:42 |
| mal | https://github.com/mer-hybris/droid-config-sony-murray/blob/master/sparse/etc/mce/70-battery-murray.ini | 23:42 |
| Mister_Magister | sure | 23:42 |
| mal | that has a different one | 23:42 |
| Mister_Magister | doesn't exist :( | 23:43 |
| mal | so look for battery related things in sysfs | 23:43 |
| mal | or charging | 23:43 |
| Mister_Magister | already ahead of you | 23:44 |
| Mister_Magister | https://paste.opensuse.org/pastes/6bf6324b1757 | 23:44 |
| Mister_Magister | i don't think it has that functionality | 23:47 |
| mal | didn't the dc sysfs have input_current_limit 1, try echo 0 there | 23:47 |
| mal | just looking at that paste and it has charging enabled stuff in discrete-charging | 23:47 |
| Mister_Magister | by monitoring battery/current_now i can see it didn't work, it's charging | 23:48 |
| Mister_Magister | it does | 23:49 |
| mal | it does what? | 23:49 |
| Mister_Magister | just looking at that paste and it has charging enabled stuff in discrete-charging> it does | 23:50 |
| mal | one option is to look at kernel sources | 23:50 |
| Mister_Magister | yeah i think we'll worry about this minor issue later | 23:50 |
| Mister_Magister | and i have to really commit all the changes or i'll forget half the fixes | 23:51 |
| Mister_Magister | enjoy your movie | 23:51 |
| mal | and about test nfc, if you don't have some tag, some bt headphones have nfc tag to pairing, also maybe some nfc bus card or something could be used | 23:53 |
| Mister_Magister | I like to enjoy audio so sadly I don't have any bluetooth headphones | 23:54 |
Generated by irclog2html.py 2.17.1 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!
