| genericv3 | hello, now i'll add most of modifications we made to sparse. I remember these right now: Fix system mount, json thingy and symlink, mask usb moded by default, libandroidicu to /odm, SELinux config & policy from vendor | 14:57 |
|---|---|---|
| genericv3 | Then i'll check irc logs to see what we done | 14:57 |
| genericv2 | i'll do them now but, how can i? | 16:12 |
| genericv2 | Can i directly add those to sparse by creating the folders or there is another way? | 16:12 |
| genericv2 | For system mount* | 16:19 |
| genericv2 | GUYS | 16:32 |
| genericv2 | TEST HWCOMPOSER GAVE DISPLAY OUTPUT NOW | 16:32 |
| genericv2 | AAAAAA | 16:32 |
| genericv2 | btw, there wa a wrong sepolicy file. i copied that with the real one and boom | 16:33 |
| genericv2 | i never thought i'll be happy for seeing this crappy animation | 16:33 |
| T42_ | <b100dian> :D | 16:34 |
| mal | genericv2: just as we suspected, some issue with selinux | 16:35 |
| mal | if test_hwcomposer gives output then getting UI working should not be too difficult | 16:35 |
| genericv2 | yeah, my stupidness level is another level for not seeing that | 16:36 |
| genericv2 | i could able to run lineage boot animation too, we fixed display thingies | 16:38 |
| genericv2 | But it's not trying to start lipstick right now, i think | 16:38 |
| genericv2 | system-etc-ld.config.28.txt.mount loaded failed failed | 16:39 |
| genericv2 | hm | 16:39 |
| genericv2 | i'll check this one | 16:39 |
| genericv2 | everything else are okay | 16:39 |
| T42_ | <elros34> impossible, lipstick is up and happy? | 16:40 |
| genericv2 | Nope, it's not even trying to start | 16:41 |
| genericv2 | but i don't care because WE GOT A SIGN OF LIFE (except telnet ofc) | 16:41 |
| genericv2 | i'll try things to make it try to start lipstick now | 16:41 |
| T42_ | <elros34> it should always start automatically | 16:42 |
| T42_ | <elros34> you can't even mask it just like that | 16:42 |
| genericv2 | some service keeps restarting and not letting it start i guess | 16:42 |
| genericv2 | Maybe these "Could not find" dudes from droid-hal-init block boot | 16:46 |
| genericv2 | mal: binder-list returns a cool enough list now | 16:51 |
| mal | good | 16:58 |
| mal | just show logs again | 16:59 |
| genericv2 | took them now, gonna reboot to twrp to pull them | 16:59 |
| T42_ | <elros34> what do you need twrp for? | 17:00 |
| genericv2 | does telnet have file pulling? | 17:00 |
| mal | is ssh working to the device? | 17:00 |
| genericv2 | ah, never tried | 17:00 |
| T42_ | <elros34> about telnet no, but its not need. you can just select and copy | 17:00 |
| mal | just set password for defaultuser from terminal first | 17:00 |
| genericv2 | set a password, also enabled pass auth . it still gives permission denied | 17:07 |
| T42_ | <b100dian> You can reverse ssh to your computer, from telnet. Actually scp | 17:20 |
| T42_ | <elros34> or use some other exotic way to copy files from telnet but why even bother when you can select logs with mouse | 17:22 |
| genericv2 | nevermind i found the issue. i can't believe myself | 17:24 |
| genericv2 | i am in, finally | 17:25 |
| T42_ | <b100dian> What was the issue? | 17:33 |
| genericv2 | me trying to ssh to the laptop itself from the laptop instead of the phone. | 17:33 |
| genericv2 | Xd | 17:33 |
| genericv2 | https://paste.opensuse.org/068d14d7e649 | 17:35 |
| genericv2 | and logcat: https://paste.opensuse.org/ed94dbede705 | 17:35 |
| T42_ | <elros34> I know reading long hadk is something people never do but please at least follow advice from it and disable audit. Look at these logs, they are filed with irrelevant spam, you will never find error in 10000 lines long text unless you know excatly what to search | 17:43 |
| genericv2 | i try to read hadk as long as i can, i forget about hadk-hot today, sorry | 17:45 |
| T42_ | <elros34> about main issue, didn't you maks user@100000? | 17:46 |
| T42_ | <elros34> mask* | 17:46 |
| genericv2 | it's not masked right now, do i need to mask it? | 17:47 |
| T42_ | <elros34> no, I do not see it that is why I ask, maybe check systemctl | 17:48 |
| genericv2 | EVERYTHING is okay except "systemd-tmpfiles-setup.service" | 17:49 |
| genericv2 | disabled audit, i'll give new logs | 17:50 |
| T42_ | <elros34> that one is probably not very relevant, I guess you may missing some cgroup configs | 17:51 |
| genericv2 | https://paste.opensuse.org/84d35cd4473b | 18:00 |
| genericv2 | https://paste.opensuse.org/44cb7c0d7f31 | 18:01 |
| genericv2 | with audit=0, idk if something changed or not | 18:01 |
| T42_ | <elros34> are you sure systemctl looks fine? For me at least wait_for_keymaster.service is incorrect | 18:13 |
| genericv2 | systemctl does not have any failing services except that unimportant tmpfiles thingy | 18:15 |
| genericv2 | wait_for_keymaster is a mess right now, but it does not appear in systemctl idk why | 18:15 |
| mal | run "binder-list -d /dev/hwbinder | grep android.hardware.keymaster" | 18:16 |
| T42_ | <elros34> fix service file and see if that will help. I Never know how to debug systemd but for sure i waits for something | 18:16 |
| genericv2 | 3.0 exist | 18:16 |
| genericv2 | android.hardware.keymaster@3.0::IKeymasterDevice/default | 18:16 |
| T42_ | <elros34> but mistake is in service file not script | 18:16 |
| T42_ | <elros34> right in script too | 18:18 |
| genericv2 | it's searching for 4.1 | 18:19 |
| genericv2 | i made it 3.0 now, i'll revert it back to 4.1 again if i am wrong | 18:19 |
| genericv2 | looks okay now, fingerprint daemon thingy launched now | 18:22 |
| genericv2 | https://paste.opensuse.org/6fbd94676c7e | 18:26 |
| genericv2 | https://paste.opensuse.org/cdd12e45fb3f | 18:26 |
| mal | so add the custom wait_for_keymaster to sparse so it overrides the one from śubmodule | 18:27 |
| genericv2 | yeah i must do that before forgetting | 18:27 |
| mal | should also systemctl output so we see which services are running and which waiting for something | 18:27 |
| T42_ | <elros34> do you watch logs continuously? I guess there should be some 1 or 2 minute timeout | 18:30 |
| genericv2 | yeah i normally do, but today didn't checked logs for that much time | 18:31 |
| genericv2 | what do i need to look for | 18:31 |
| genericv2 | mal: took them, now sending | 18:32 |
| mal | is user session running? | 18:32 |
| genericv2 | Like default.target? | 18:32 |
| mal | "systemctl-user" or "systemctl --user" | 18:32 |
| T42_ | <elros34> user@100000 is not started | 18:32 |
| genericv2 | Failed to connect to bus: No such file or directory | 18:35 |
| genericv2 | how, it's not masked or anything | 18:35 |
| genericv2 | i'll check | 18:35 |
| T42_ | <elros34> systemctl status user@100000 | 18:35 |
| genericv2 | loaded but dead | 18:36 |
| genericv2 | hm | 18:36 |
| T42_ | <elros34> can you show you droid-config? | 18:37 |
| genericv2 | user@100000 was working yesterday, very strange | 18:38 |
| T42_ | <elros34> maybe also systemctl list-dependencies | 18:38 |
| mal | just show the systemctl output, I know some ways things could break | 18:40 |
| genericv2 | https://paste.opensuse.org/pastes/47b7cd2d2756 | 18:40 |
| genericv2 | oh, services failed after i checked systemctl output. now i see them | 18:45 |
| genericv2 | droid-bootctl etc is dead | 18:54 |
| genericv2 | mal: any idea or its my fault? I'll mess with it if i made a stupid thing again. | 19:00 |
| mal | genericv2: that says now that user@100000 is running | 19:05 |
| mal | try again systemctl-user | 19:05 |
| genericv2 | i accidently uploaded the wrong file | 19:11 |
| genericv2 | that was a previous output of systemctl | 19:12 |
| genericv2 | mal: https://paste.opensuse.org/74fc2103f4b0 | 19:13 |
| genericv2 | Thanks for being patient for my stupid mistakes | 19:14 |
| mal | do you have gmp-droid installed? | 19:18 |
| mal | I have a feeling this could be same issue I saw on one other android 13 based device during early porting | 19:19 |
| genericv2 | https://paste.opensuse.org/pastes/6751529349e0 it is, i guess... | 19:20 |
| mal | run "zypper rm gmp-droid" | 19:20 |
| genericv2 | it's installed | 19:21 |
| mal | yes, let's uninstall it for now | 19:21 |
| genericv2 | it'll uninstall these aswell: droid-hal-version-sagit patterns-sailfish-device-adaptation-sagit patterns-sailfish-device-configuration-sagit | 19:21 |
| genericv2 | looks scary :D | 19:22 |
| mal | does it allow breaking dependencies? | 19:22 |
| mal | usually it gives option for that | 19:22 |
| genericv2 | hmm, no i think. | 19:22 |
| genericv2 | I only can choose y n and v | 19:23 |
| genericv2 | ? gives more options but none of them allows to break dependencies | 19:23 |
| mal | try "zypper rm -R gmp-droid" | 19:23 |
| genericv2 | still same | 19:24 |
| mal | hmm | 19:24 |
| genericv2 | looks bad | 19:24 |
| T42_ | <elros34> rpm --nodeps or remove files rpm -ql gmp-droid | 19:25 |
| mal | we want to get rid of the oneshot that package installs | 19:26 |
| mal | it can hang forever is something is wrong in media side | 19:26 |
| mal | *if | 19:26 |
| genericv2 | i made something: https://paste.opensuse.org/pastes/de4d14e3b61b | 19:28 |
| T42_ | <elros34> rpm -e --nodeps gmp-droid | 19:30 |
| genericv2 | it's gone now | 19:31 |
| genericv2 | 'gmp-droid' not found in package names. | 19:31 |
| genericv2 | (the second one is after trying zypper rm to confirm it's removed) | 19:31 |
| genericv2 | now it tried to start things at least, but failed with strange errors | 19:36 |
| genericv2 | droid-hal-init is broken again | 19:37 |
| genericv2 | oh no, not this again | 19:39 |
| genericv2 | libc: Unable to set property "droid.late_start" to "trigger_late_start": connection failed; errno=111 (Connection refused) | 19:39 |
| genericv2 | https://paste.opensuse.org/pastes/f2068b4f7405 | 19:41 |
| genericv2 | oh no, binders broke again | 19:43 |
| mal | any chance droid-hal-init times out? | 19:43 |
| genericv2 | maybe, i'll try to increase it's timeout value | 19:45 |
| genericv2 | it also failed with timeout 60 seconfs | 19:48 |
| genericv2 | will logcat be enough? | 19:48 |
| mal | show both journal and logcat | 19:50 |
| genericv2 | i rebooted the laptop instead of the phone. connecting the thing to the wifi is another pain. | 19:51 |
| mal | take the journal after droid-hal has timed out so we see how it breaks | 19:51 |
| mal | if logcat works keep it running until it fails | 19:51 |
| genericv2 | https://paste.opensuse.org/ebeb34043681 and https://paste.opensuse.org/e54d5c777a5a | 19:54 |
| mal | not sure why it fails now | 20:16 |
| mal | did you do anything else since it was working before | 20:16 |
| genericv3 | i don't know, i'll think about it. Lets create a list of our known-working modifications. I added some of them to sparse but it's not done yet, maybe a new package can fix my issues. | 20:28 |
| genericv3 | Also if something comes to your mind about this please write it with tagging me, i am checking irc logs for these :) | 20:29 |
| genericv3 | And as always bye, at least we got a graphical sign of life today :D | 20:30 |
| T42_ | <elros34> not list but simply upload droid-config to github | 20:40 |
Generated by irclog2html.py 2.17.1 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!