Friday, 2024-03-08

Mister_Magisterman getting back into the loop is always painful10:21
Mister_Magisterit's like "come on you know that" and you search for info you know that you know but you forgot10:21
Mister_Magistermal: are you travelling/busy?10:32
* Mister_Magister assume yes10:57
Mister_Magisteri'm doing some porting and was wondering if i can bother you11:18
malhow about you just ask the questions here and me or someone else can answer11:22
* Mister_Magister mal's upset run11:24
malno, just the usual thing that just ask, don't ask about asking11:25
Mister_Magisteryee i know11:26
Mister_Magisterif anyone doesn't mind having a look about what don't i need anymore, i yeeted systemsettings and bluebinder so far
malI might have a look after lunch11:41
Mister_Magisteri'm updating the port from 3.3 armv7hl to 4.5 aarch6411:42
Mister_Magistera lot to do a lot can go wrong11:43
Mister_Magisterand for whatever reason i'm getting nothing provides pkgconfig(android-headers), nothing provides community-adaptation which is very weird11:45
Mister_Magistergotta go for a sec but will be in an hour till the end of the day11:47
Mister_Magisterand locally i'm getting this and i also don't know why
Mister_Magisteram back12:43
Mister_Magisterooh i lacked _aarch6412:52
Mister_Magisteri'm basically going package by package and checking if its in adaptation-community, if yes, yeet12:58
Mister_Magistermal: do you mind creating nemo:testing:hw:oneplus:fajita? i don't think both devices can be in same project at least it's beyond my skill13:09
Mister_Magisterahh i have armv7hl droid-hal13:17
Mister_Magisterkimmoli: ping can you tell me what is this for?
Mister_Magisterohhhhhh it's the button13:31
Mister_Magisternow that makes a ton of sense13:31
Mister_Magister@elros34 ye i know but it wasn't helping xd13:32
Mister_Magisteranyone know why pulseaudio refuses to build? error: 'AUDIO_DEVICE_IN_HDMI_ARC' undeclared here (not in a function);13:40
Mister_Magister yeah i don't think droid base 11 will work with my base 913:53
Mister_Magisterahh i added repo -jb2q but didn't change repo in _service13:56
Mister_Magisteri need -jb2q don't i13:56
Mister_Magistereverything built so now i just need to build zip and see it fail14:05
Mister_Magisterit doesn't boot and i get no telnet :/ but it does get past initram15:21
Mister_Magistermal: quick question, can i revert patches?15:24
Mister_Magisterprobably need to yeet everything and resync don't i15:26
malwhat patches?15:26
malnot sure if there have been fixes in your branch, check git diff15:27
Mister_Magisterthere have been couple of them15:27
Mister_Magisterquite a lot15:27
Mister_Magisteri'll just rm -rf * and resync15:28
T42_<elros34> I guess you had no telnet because you have build testing image instead devel one16:19
Mister_Magisterdoes that matter?16:28
Mister_Magisterall my devices are built against testing and they all have telnet16:28
Mister_Magisterand no i didn't have the no_telnet file16:28
Mister_Magisteraccording to init.log it was trying to enable rndis but there was no rndis action whatsoever16:28
maldo you have correct kernel config options?16:44
maltesting images don't have telnet in sfos side16:45
Mister_Magisterthat's weird then that all my other devices do have telnet, did it change recently?16:45
Mister_Magisteri didn't touch kernel really16:46
Mister_Magisterthere is CONFIG_USB_CONFIGFS_RNDIS=y16:46
Mister_Magisterthis time i built zip from local files and there's still no telnet16:49
Mister_Magisterwas thinking the updated patches would help but nah16:50
maldid that have telnet before?16:50
malwith old build16:51
maldo you have CONFIG_USB_CONFIGFS_RMNET_BAM=y ?16:51
Mister_Magistermal: it did on the previous author's 3.3 build16:51
Mister_MagisterI do not have CONFIG_USB_CONFIGFS_RMNET_BAM even as an option16:52
T42_<elros34> init_disable_telnet for testing images is something which exist since always assuming you use testing common repo17:19
Mister_Magisteri literally told you already i don't have that file17:20
Mister_Magisterliterally first thing i checked for :P17:22
Mister_Magisterif i don't get telnet i'll probably enable persistant journal and check it from chroot17:23
Mister_Magisteras i don't have any better ideas17:24
T42_<elros34> and I literally just answer your question whether this is something which changed recently17:24
Mister_Magisterye so it didn't change and telnet is still in testing17:24
Mister_Magistereh still nothing, i tired updating hybris-boot17:26
T42_<elros34> did you try init_enter_debug2 assuming your init.log errors ar not fatal?17:26
Mister_Magisteri tried but without telnet it doesn't help17:27
Mister_Magisteri get the mtp device so its post ramdisk for sure17:29
Mister_Magisterhere's init.log
Mister_Magisteri can see the bam stuff but my kernel doesn't have it17:30
Mister_Magisteralso init-debug-stderrout has one line + exec17:31
T42_<elros34> init.log looks correct, init_enter_debug2 should do the job17:33
Mister_Magisterthat just makes it hang on the mtp device mate17:35
Mister_Magister there's no rndis whatsoever17:35
Mister_Magisterusb 1-8: SerialNumber: Mer Debug telnet on port 2323 on rndis0 - also running udhcpd17:37
Mister_Magisterit's saying its running it but  there's none17:37
T42_<elros34> ifconfig -a?17:37
Mister_Magisterthere's no new device in dmesg mate ofc there's no network device either17:38
Mister_Magister@elros34 i did sanity check and plugged in my zenfone 8… no rndis either… i… went to check on laptop. THERE IS RNDIS17:42
Mister_Magistergoddamnit xd17:43
Mister_Magisterit's host issue17:43
Mister_Magistermodprobing rndis_host worked17:44
Mister_Magisteraltho dhcp doesn't work that's no biggie17:44
Mister_Magisteri'm in17:45
Mister_Magisterhm that's weird, despite building 4.5 image i don't have defaultuser17:47
maldid you renegerate .ks?17:49
Mister_Magisteryes i did17:50
Mister_Magisterthere's no mention of defaultuser nor nemo in .ks though17:50
Mister_Magisterthere's no nemo nor defaultuser user either17:51
Mister_Magisterthe only culprit i can think of is this
Mister_Magisterbut that should be post build17:52
malthe user should be created at boot afaik17:52
Mister_Magisterwell they're not17:53
Mister_Magisteri can see the oneshots in .ks file if thats what you mean17:54
T42_<elros34> did you rebuild rpm cache when using local repo, its now needed17:55
Mister_Magisteri probably did not? how do i do that17:56
Mister_Magisteri did not have any old build either this is all fresh17:58
T42_<elros34> so probably not and issue, you can check whether you have matching versions in device and local repo. Just for  the record I do not remember exactly, maybe this was enough:
T42_<elros34> user name is set in /usr/lib/startup/start-autologin18:00
Mister_Magister@elros34 i don't think so, it just sets defaultuser for login18:05
Mister_Magisterit doesn't create defaultuser18:05
Mister_Magisterjust grabs the 1000018:05
Mister_Magisterbut the autologin@100000.service does not exist18:07
Mister_Magisterit was never ran18:07
Mister_Magisterwhich is probalby becaues it never found the use18:08
Mister_Magisteroh ye user@0.service18:08
Mister_Magister this is current status18:09
Mister_Magistersomething is definitely wrong with the user session and I think that's blocking me currently as init-done is waiting for user session and it never happens18:10
T42_<elros34> you have bunch of systemd ints in old/wrong path
Mister_Magisteri'll move them to usr18:16
Mister_Magisteryeah bootctl is not running and neither is vendor18:20
Mister_Magisterno good18:20
Mister_Magisterthanks @elros34!18:20
Mister_Magisteri feel out of the loop hard :P18:23
T42_<elros34> you may look at release specific issues:
malbootctl is not needed on all devices18:30
Mister_Magisterwell it is on this one18:30
Mister_Magisterthere's not much for 4.518:30
Mister_Magisteri did all the needed previous changes i just missed the lib18:31
Mister_Magisterok so with moved the systemd services the services are now here but still no dice18:34
Mister_Magister this is current status18:35
Mister_Magisterstill no defaultuser18:36
T42_<elros34> any timeouts after about 2min in journal?18:37
Mister_Magisterby timeouts you mean…?18:38
Mister_Magisterconnman is dying in the loop18:39
T42_<elros34> some services are waiting, then they should timeout and die/restart, this should be logged18:39
T42_<elros34> maybe remove that custom connman.service18:40
Mister_Magisterwhat custom? its not custom18:42
Mister_Magisternothing is waiting, the only waiting service is the init-done which is waiting for user session that's never happening18:43
Mister_Magisterbecause there's no user18:43
T42_<elros34> its 4 years old, some lines changed, it shouldn't be created like that.
Mister_Magisterlol i didn't see it18:44
Mister_Magisterahh its in /etc18:44
Mister_Magisterwhaddya know we got defaultuser now18:47
Mister_Magisterfreaking magic18:47
Mister_Magisterstill no gui but heck, its progress18:48
Mister_Magisterye we got user@100000.service and stuff18:48
Mister_Magister@elros34 you deserve a smooch18:48
Mister_Magisterokay other than fingerprint everything is running18:48
T42_<elros34> if it still requirest that fix from custom connman.service then use 'systemctl edit connman.service' method intead18:49
Mister_Magisterye i'll update it if needed18:49
Mister_Magisterfor now i gotta figure out why lipstick is dead18:50
Mister_Magistersurfaceflinger cleared screen so there's that18:50
Mister_Magister /usr/sbin/dummy_compositor: unrecognized option '--hwc-restart' i know that one18:51
T42_<elros34> it will bite until  sailfish 4.6.0.x18:52
Mister_Magisterhow was it being fixed?18:53
Mister_Magisterit's definitely fixed on my other devices but i do'nt remember how xd18:53
Mister_Magisteri remember the issue i don't remember how it was solved xd18:53
T42_<elros34> build latest mce or downgrade some submodule18:53
T42_<elros34> or remove some files from systemd18:54
Mister_Magisteryeah it was downgrading some submodule18:54
Mister_Magisterwill use sake's commit18:55
Mister_Magisteroh? WE GOT UI18:58
Mister_Magisterthat was easy18:58
Mister_Magister@elros34 i would smooch you if i could18:58
T42_<elros34> glad you can't :P19:00
Mister_Magisteri just need slight push when i'm stuck and i'll be right back in the loop :P19:00
Mister_Magisterit's like muscle memory, i'm just rusty xd19:01
T42_<Mister_Magister> yo @adampigg do you mind pointing me into direction of the CI/CD for building sfos images cause my jenkins script doesn't work at all and i'm tired of debugging it19:02
T42_<elros34> not sure if this is what you ask about
T42_<adampigg> i have another one in github19:11
T42_<adampigg> so, there is porters-ci on gitlab, but gitlab changed terms to reduce number of hours available19:13
T42_<adampigg> so, there is this also
T42_<Mister_Magister> thanks19:15
Mister_Magisterbtw any clue why i'm getting /usr/lib/python3.8/site-packages/pykickstart/commands/ KickstartParseWarning: A repo with the name adaptation-community-fajita- has already been defined.?19:29
Mister_Magisterit prevents me from using the obs repo19:29
T42_<b100dian> @adampigg we really should make a gh action but never had the time, so all 'round copies it is:-S19:39
T42_<adampigg> yes, this is true19:39
T42_<NotKit> another copy, don't remember why, but vidofnir had something specific to building super.img20:51
T42_<adampigg> 😂20:51
T42_<b100dian> Last one to copy makes GH action :p21:01
T42_<Mister_Magister> funnily enough enchilada is already on gitlab ci22:00
Mister_Magisterguys… there was issue where sailjail wasn't working22:07
Mister_Magisteri know elros knows what i'm talking about22:07
Mister_Magisteri think there was something in kernel22:07
Mister_Magister@elros34 pardon the ping but i'm sure you know what i'm talking about22:11
Mister_Magister i think this might be it… it's even the same kernel22:15
Mister_Magisterno that wasn't it22:27
Mister_Magisteri'm getting weird stuff from sailjail Effective group is not privileged\n", 93../daemon/sailjailclient.c:1478: sailjailclient_main(): W: Effective group is not privileged22:43
Mister_Magisternevermind fixed
Mister_Magisterthanks @elros3422:51
T42_<Pachof> It is normal that it takes a while to compile in this part "out/soong/.bootstrap/bin/soong_build out/soong/"?23:14
T42_<elros34> @Mister_Magister no idea about error. Does dummy-saijail even work in 4.5.0.x? Notkit made fork and updated it for 4.5.0, check it23:38
Mister_Magisteridk i used your repo and it works so…23:39
mal@Pachof how long does it take?23:44
T42_<Pachof> 190 minutos and 27 secs23:46
malthat seems quite excessive23:46
T42_<Pachof> 190 minutes and 27 secs (edited)23:46
malhow much ram do you have?23:46
T42_<Pachof> 8gb ram + 16 GB swap23:47
malhow much is free now?23:47
maldoes the process show in top?23:47
T42_<Pachof> How do I look at that?23:47
malrun "top" in terminal23:47
Mister_Magisterthats not a lot of ram23:48
T42_<Pachof> At the top is soong_build and the free physical memory is 362MB and in swap it is free 8164MB23:49
malbut it uses some swap at least?23:50
malI remember I had some issues in the past to use swap in android builds, not sure how well that works now23:50
T42_<Pachof> Yes, yes I would believe that they are using swap memory because they have not sent any out of memory error (re @SailfishFreenodeIRCBridgeBot: <mal>but it uses som...)23:51
T42_<elros34> if it hangs for 3h than don't wait, restart. Did you enable swap before starting building?23:51
malyeah, no point in waiting even that long23:52
T42_<Pachof> The swap is enabled since I installed my Linux (re @elros34: if it hangs for 3h t...)23:53

Generated by 2.17.1 by Marius Gedminas - find it at!