| CarbontatedCaffeine | Is this a ngfd problem? | 00:00 |
|---|---|---|
| mal | probably many issues | 00:07 |
| mal | ngfd and some others can be probably fixed by something like https://github.com/mlehtima/droid-config-fp3/tree/master/sparse/odm the symlinks in lib and lib64 folders are probably needed, you need to find the targets for the symlinks on your device | 00:08 |
| CarbontatedCaffeine | Ok! | 00:09 |
| mal | did you yet check what the vendor partition contains on your device? | 00:09 |
| CarbontatedCaffeine | The targets being, the libraries? | 00:10 |
| CarbontatedCaffeine | "did you yet check what the vendor partition contains on your device?" Uhh it isn't mounted | 00:10 |
| mal | like I said before, try mounting it manually and see what it contains | 00:10 |
| CarbontatedCaffeine | I still don't know whether I need it or not | 00:10 |
| CarbontatedCaffeine | Okay | 00:11 |
| mal | that is why I asked you to check what it contains to see if it's needed or not | 00:11 |
| mal | you need to find where the actual file in those symlinks are on your device | 00:11 |
| mal | so for example "find / -name libandroidicu.so" | 00:11 |
| CarbontatedCaffeine | "/system_root # mount /dev/block/bootdevice/by-name/vendor_a ./vendor/ mount: /system_root/vendor: wrong fs type, bad option, bad superblock on /dev/sda31, missing codepage or helper program, or other error. dmesg(1) may have more information after failed mount system call." | 00:12 |
| CarbontatedCaffeine | "/ # find / -name libandroidicu.so find: /sys/kernel/slab/L2TP/IPv6: No such file or directory find: /sys/kernel/slab/L2TP/IP: No such file or directory /system/apex/com.android.runtime.release/lib64/libandroidicu.so /system/apex/com.android.runtime.release/lib/libandroidicu.so /system_root/system/apex/com.android.runtime.release/lib64/libandroidicu.so /system_root/system/apex/com.android.runtime.release/lib/libandroidicu.so" | 00:13 |
| mal | so you have the target in that list, the system one, of course pick the correct lib vs lib64 to correct symlink | 00:14 |
| CarbontatedCaffeine | uhm | 00:15 |
| CarbontatedCaffeine | Sorry, can you explain that to me a little simplier | 00:16 |
| CarbontatedCaffeine | like how do I symlink these two, and how do I make the changes stick using the source code. | 00:17 |
| mal | "mkdir -p /odm/lib64" then "cd /odm/lib64" then "ln -s /system/apex/com.android.runtime.release/lib64/libandroidicu.so" | 00:18 |
| mal | similar to the other symlinks but different filenames | 00:18 |
| mal | you can create the symlinks in correct folders in sparse as usual | 00:19 |
| mal | later when you have first tested those on device | 00:19 |
| mal | of course when creating those in sparse don't use absolute path when doing that cd /odm/lib64 | 00:19 |
| mal | which should be obvious | 00:19 |
| CarbontatedCaffeine | Oh, it's that easy? | 00:31 |
| CarbontatedCaffeine | huh | 00:31 |
| CarbontatedCaffeine | cool | 00:31 |
| CarbontatedCaffeine | What does sparse mean? | 00:32 |
| mal | hybris/droid-configs/sparse folder in your sources | 00:39 |
| mal | it's where all kinds of configuration and other stuff is added | 00:39 |
| mal | it's part of the droid-configs repo | 00:39 |
| CarbontatedCaffeine | Ah, cool | 00:40 |
| mal | I'll go offline now | 00:41 |
| CarbontatedCaffeine | I'm looking through my journal, but I can't find lipstick, thanks for your help | 00:41 |
| CarbontatedCaffeine | have a good one | 00:41 |
| mal | you might need libui_compat_layer which is an optional library built in habuild_sdk and packaged in droid-hal | 00:43 |
| mal | since log says it can't find gralloc and that is the way for devices without gralloc library | 00:44 |
| mal | so you can test if that helps with lipstick issue, so "make libui_compat_layer" in habuild_sdk and then the usual build_packages.sh -d | 00:45 |
| CarbontatedCaffeine | cheers | 00:45 |
| mal | but first test the symlinks for the other issues seen in log like ngfd | 00:45 |
| CarbontatedCaffeine | the journal is extremely large | 00:46 |
| mal | yeah | 00:46 |
| CarbontatedCaffeine | a lot of things seem to be working now | 00:46 |
| CarbontatedCaffeine | The status LED is now stuck white | 00:46 |
| CarbontatedCaffeine | which I think is good | 00:46 |
| mal | you can check what is working and what is not by checking output of "systemctl" and "systemctl-user" | 00:46 |
| CarbontatedCaffeine | oh cool | 00:46 |
| mal | latter can also be systemctl --user depending on which environment you use, I think telnet needs systemctl-user | 00:47 |
| CarbontatedCaffeine | Failed to connect to bus: No such file or directory | 00:47 |
| CarbontatedCaffeine | both give me this | 00:48 |
| mal | ok, maybe user session is not running | 00:48 |
| mal | you mean also for just systemctl? | 00:48 |
| CarbontatedCaffeine | yes | 00:48 |
| CarbontatedCaffeine | no | 00:48 |
| CarbontatedCaffeine | wait | 00:48 |
| CarbontatedCaffeine | systemctl works fine | 00:48 |
| CarbontatedCaffeine | Would you like to see the journal | 00:48 |
| mal | it tells in there is user session is running | 00:49 |
| mal | well I won't be looking at logs for quite many hours so maybe try checking what is failing and test the things I mentioned | 00:49 |
| CarbontatedCaffeine | Alright, thanks | 00:49 |
| mal | and checking channel logs for errors can also help | 00:50 |
| mal | you can search channel logs here https://piggz.co.uk/sailfishos-porters-archive/ | 00:50 |
| CarbontatedCaffeine | Even better | 00:53 |
| CarbontatedCaffeine | I've been using that actually | 00:53 |
| CarbontatedCaffeine | very good, sometimes | 00:53 |
| CarbontatedCaffeine | lots of reading to do then | 00:54 |
| CarbontatedCaffeine | Hey everyone, I have some great progress to report to you | 03:20 |
| CarbontatedCaffeine | I have /vendor mounted now | 03:20 |
| CarbontatedCaffeine | I also added the symlink to libandroidicu.so for odm/lib64 to sparse in droid config | 03:21 |
| CarbontatedCaffeine | which worked out of the box | 03:21 |
| CarbontatedCaffeine | Is there anyone currently online that may be able to help debug, thanks. | 04:06 |
| CarbontatedCaffeine | https://pastebin.com/pbXtyrih | 04:30 |
| T42 | <Mister_Magister> @b100dian how's your port coming along? | 06:28 |
| T42 | <b100dian> Debugging the fingerprint sensor. I have in my todo also: wifi drivwr consumes more battery and video playback is interrupted with a hardware error. But it should boot, have all these if I make a new install image https://github.com/sailfishos-on-sake/Release/releases/tag/release-devel-4.5.0.19-alpha-1.6, plus video recording | 07:41 |
| T42 | <b100dian> Also I have trouble booting twrp currently so manual installation may be needed if thats also your case too https://github.com/sailfishos-on-sake#alternate-installation-instructions | 07:46 |
| T42 | <b100dian> Nfc, Mtp, Fm radio, actdead, and other device specifics like 90Hz display is not yet looked into. | 07:50 |
| CarbontatedCaffeine | Hello, might anyone know what is wrong here, I feel stuck currently: https://pastebin.com/pbXtyrih | 11:41 |
| CarbontatedCaffeine | I don't have a GUI yet | 11:41 |
| mal | you need to provide journal log | 11:42 |
| CarbontatedCaffeine | Of course | 11:42 |
| CarbontatedCaffeine | https://pastebin.com/F3DU9SNm | 11:54 |
| CarbontatedCaffeine | here is journal | 11:55 |
| mal | something went really wrong, droid-hal-init is not even starting | 12:10 |
| mal | which android base is that? I forgot | 12:11 |
| CarbontatedCaffeine | 11 | 12:22 |
| CarbontatedCaffeine | May I say that droid-hal-init hasn't ever started | 12:23 |
| CarbontatedCaffeine | I saw some people in logs say you didn't need it? But I was skeptical | 12:23 |
| mal | that one of the most critical processes | 12:24 |
| mal | no way to get UI working without | 12:24 |
| CarbontatedCaffeine | right | 12:25 |
| mal | do you have that libc.so somewhere on device? | 12:26 |
| CarbontatedCaffeine | Give me a tick to check | 12:26 |
| CarbontatedCaffeine | https://pastebin.com/KysDc1X5 | 12:27 |
| CarbontatedCaffeine | in many places, as seen here | 12:28 |
| CarbontatedCaffeine | do I need to make another symlink? | 12:28 |
| mal | no symlink should be needed for that | 12:32 |
| CarbontatedCaffeine | Oh | 12:33 |
| mal | you probably need to strace droid-hal-init | 12:33 |
| mal | is for example /system/lib64/libc.so normal file? | 12:33 |
| CarbontatedCaffeine | EXECUTABLE "droid-hal-init": library "libm.so" not found | 12:34 |
| mal | ? | 12:34 |
| mal | I said no symlinks for those, the issue is something else | 12:35 |
| CarbontatedCaffeine | sorry | 12:35 |
| CarbontatedCaffeine | It was before you responded | 12:36 |
| CarbontatedCaffeine | I was merely curious | 12:36 |
| mal | that is why I asked you to strace droid-hal-init | 12:37 |
| CarbontatedCaffeine | I did | 12:37 |
| mal | where is the output then | 12:37 |
| CarbontatedCaffeine | Okay, I removed my symlink | 12:37 |
| CarbontatedCaffeine | and rebooted | 12:38 |
| CarbontatedCaffeine | heres the output | 12:38 |
| CarbontatedCaffeine | https://pastebin.com/HmrG5pvr | 12:38 |
| mal | is /system/lib64/libc.so normal file? | 12:39 |
| CarbontatedCaffeine | "/system/lib64/libc.so" is a symlink | 12:39 |
| CarbontatedCaffeine | I can confirm | 12:39 |
| CarbontatedCaffeine | using "ls" shows it as cyan,' | 12:40 |
| mal | where does it point? | 12:40 |
| mal | use ls -l | 12:40 |
| T42 | <k1gen> didn't enter this room for a while, and when I decide to check on it, I see mal doing God's work here again, helping some guy :D | 12:40 |
| CarbontatedCaffeine | "/ # readlink -f /system/lib64/libc.so /system/lib64/libc.so" | 12:41 |
| T42 | <k1gen> CarbontatedCaffeine: you know, I was in your place couple months ago :) | 12:41 |
| CarbontatedCaffeine | What the hell :skull: | 12:41 |
| CarbontatedCaffeine | IT'S A SYMLINK TO A SYMLINK | 12:41 |
| CarbontatedCaffeine | paradox | 12:41 |
| CarbontatedCaffeine | "CarbontatedCaffeine: you know, I was in your place couple months ago :)" If you mean having mental breakdowns every couple of hours | 12:41 |
| CarbontatedCaffeine | I have been working on this point, every day, for a WEEK | 12:42 |
| CarbontatedCaffeine | *port | 12:42 |
| T42 | <k1gen> no, I mean being constantly helped by mal and elros :) | 12:42 |
| CarbontatedCaffeine | 1 week anniversary today. | 12:42 |
| CarbontatedCaffeine | "no, I mean being constantly helped by mal and elros :)" I'm quite an annoyance | 12:42 |
| CarbontatedCaffeine | I know | 12:43 |
| T42 | <k1gen> I'm sure I was one too | 12:43 |
| CarbontatedCaffeine | Anyway, uhh so I gotta figure this out | 12:43 |
| CarbontatedCaffeine | which is the real libc.so | 12:43 |
| mal | CarbontatedCaffeine: it took me months to get my first device to boot to UI but that was very long time ago and things are easier now usually | 12:43 |
| T42 | <k1gen> yeah, there's also sound up ahead :) | 12:44 |
| CarbontatedCaffeine | Meh, gui will allow me to sleep easy | 12:44 |
| CarbontatedCaffeine | for a while | 12:45 |
| T42 | <k1gen> it won't, I had gui but no sound for a while | 12:45 |
| CarbontatedCaffeine | I can't find the real libc.so | 12:45 |
| CarbontatedCaffeine | they're imposters | 12:45 |
| CarbontatedCaffeine | "/system/apex/com.android.runtime.release/lib64/bionic/libc.so" is real | 12:47 |
| CarbontatedCaffeine | wait | 12:48 |
| CarbontatedCaffeine | "lrw-r--r-- 1 root root 46 Jan 1 2009 /system/lib64/libc.so -> /apex/com.android.runtime/lib64/bionic/libc.so" | 12:48 |
| CarbontatedCaffeine | oh it is just a symlink to that one | 12:48 |
| CarbontatedCaffeine | I see | 12:48 |
| mal | but different path | 12:48 |
| mal | wait | 12:48 |
| mal | run "ln -s /system/apex /apex" | 12:49 |
| CarbontatedCaffeine | why? | 12:49 |
| CarbontatedCaffeine | nevermind | 12:49 |
| CarbontatedCaffeine | I shouldn't question | 12:49 |
| mal | other path is /apex other is /system/apex | 12:49 |
| CarbontatedCaffeine | ah | 12:50 |
| mal | you see the file is in /system/apex/com.android.runtime.release/lib64/bionic/libc.so but symlink points to /apex/com.android.runtime/lib64/bionic/libc.so | 12:50 |
| CarbontatedCaffeine | Rebooted, it's still complaining about libc.so not found | 12:51 |
| CarbontatedCaffeine | darn | 12:51 |
| CarbontatedCaffeine | Oh, it's because /apex already exists | 12:52 |
| CarbontatedCaffeine | so it just places a new symlink in /apex/apex | 12:52 |
| CarbontatedCaffeine | whoops | 12:53 |
| mal | ah, what is what we want | 12:53 |
| CarbontatedCaffeine | oh | 12:53 |
| mal | so is /apex a folder or symlink? | 12:53 |
| CarbontatedCaffeine | a folder | 12:53 |
| mal | I meant that we want /apex to point to /system/apex | 12:53 |
| CarbontatedCaffeine | and the command put a symlink called apex, into that folder | 12:53 |
| CarbontatedCaffeine | yes, I know you did | 12:54 |
| mal | remove the /apex folder and create the symlink | 12:54 |
| CarbontatedCaffeine | easy | 12:54 |
| CarbontatedCaffeine | okay, done | 12:54 |
| mal | other option is to bind mount /system/apex to /apex, we'll see which way we should go | 12:54 |
| CarbontatedCaffeine | rebooting | 12:54 |
| mal | but symlink for now | 12:54 |
| CarbontatedCaffeine | both work as well as each other right? | 12:55 |
| CarbontatedCaffeine | Hold up, something isn't right | 12:55 |
| CarbontatedCaffeine | "/ # ls /apex/ com.android.conscrypt com.android.media com.android.media.swcodec com.android.resolv com.android.runtime.release com.android.tzdata" | 12:55 |
| CarbontatedCaffeine | That's not apex? | 12:56 |
| mal | wait | 12:56 |
| mal | remove the symlink and create the folder again | 12:56 |
| CarbontatedCaffeine | done | 12:57 |
| mal | then run "ln -s /system/apex/com.android.runtime.release /apex/com.android.runtime" | 12:57 |
| CarbontatedCaffeine | Done | 12:57 |
| CarbontatedCaffeine | checking again once rebooted | 12:57 |
| CarbontatedCaffeine | fixed, but new issue | 12:58 |
| CarbontatedCaffeine | https://pastebin.com/7HKRgPVF | 12:59 |
| CarbontatedCaffeine | I'm not how to explain this one | 12:59 |
| CarbontatedCaffeine | the end part | 13:00 |
| CarbontatedCaffeine | "write(2, "cannot locate symbol \"_ZN3fmt2v6"..., 123cannot locate symbol "_ZN3fmt2v68internal10basic_dataIvE23zero_or_powers_of_10_32E" referenced by "/sbin/droid-hal-init"...) = 123" | 13:00 |
| T42 | <b100dian> `fmt::v6::internal::basic_data<void>::zero_or_powers_of_10_32` says demangler.com | 15:37 |
| T42 | <b100dian> Maybe you can grep for your string to find which .so has that symbol..? | 15:52 |
| CarbontatedCaffeine | I'll give that a try thanks | 21:19 |
| CarbontatedCaffeine | it references the error is coming from "/sbin/droid-hal-init"? | 21:24 |
| CarbontatedCaffeine | I'm not sure where to go from here | 21:41 |
| CarbontatedCaffeine | https://pastebin.com/UhJvaLpX | 21:43 |
| CarbontatedCaffeine | heres the strace for droid-hal-init currently | 21:43 |
| mal | just to be sure, you have correct android base version flashed on the device before you installed sailfish? | 21:44 |
| CarbontatedCaffeine | Android 11 (Lineage) | 21:47 |
| CarbontatedCaffeine | Would it be a good idea if I re-flash it? | 21:48 |
| CarbontatedCaffeine | just in case | 21:49 |
| CarbonatedCaffeine | It bootloops forever now, I have no idea how I got it cooperate the first time | 22:19 |
| CarbonatedCaffeine | one step forward, 15 steps backwards | 22:21 |
| CarbonatedCaffeine | doing touch /data/.stowaways/sailfishos/init_enter_debug2 allowed it to boot | 22:31 |
| CarbonatedCaffeine | but not into the rootfs obviously | 22:31 |
| CarbonatedCaffeine | if I delete it again | 22:32 |
| CarbonatedCaffeine | lets see | 22:32 |
| CarbonatedCaffeine | nope | 22:32 |
| T42 | <elros34> init_enter_debug2 boots to rootfs just stop booting system | 22:33 |
| T42 | <elros34> there is even instruction shown when you telnet with thta file what to do to continue booting | 22:33 |
| CarbonatedCaffeine | Wait | 22:33 |
| CarbonatedCaffeine | Do you remember when I was having problems with that | 22:34 |
| CarbonatedCaffeine | I wasn't able to telnet when I did that | 22:34 |
| T42 | <elros34> That way of preventing system start is very usefull for bootloops because you can run journalctl -f or dmesg -w, continue booting and see what fails | 22:35 |
| CarbonatedCaffeine | elros34: I can't get into the system | 22:35 |
| CarbonatedCaffeine | like I said | 22:35 |
| T42 | <elros34> no idea what does it mean "you can't get into system" | 22:36 |
| CarbonatedCaffeine | no telnet | 22:36 |
| CarbonatedCaffeine | the problem I had last time when making that file | 22:36 |
| T42 | <elros34> yeah but you were using wrong telnet port or didn't set inet address for usb iterface. Never showed any log of what you have done so just guessing | 22:37 |
| CarbonatedCaffeine | But.. | 22:37 |
| CarbonatedCaffeine | I told you | 22:37 |
| CarbonatedCaffeine | it wasn't working | 22:37 |
| T42 | <elros34> you think this is first device with telnet issues. I had this and many before you had this too | 22:38 |
| CarbonatedCaffeine | Are you having MY issue? | 22:38 |
| T42 | <elros34> I even had some mainline which loose conection very few seconds so run this ip route/set commands in loop | 22:39 |
| CarbonatedCaffeine | okay? | 22:40 |
| CarbonatedCaffeine | It gets to the rootfs | 22:41 |
| CarbonatedCaffeine | but crashes straight after that | 22:42 |
| T42 | <elros34> the point is if you do not know how to fix something then show your logs, host dmesg from connecting device, ifconfig output, ip route.. commands and so on. "Doesn't work" is not usefull at all | 22:42 |
| CarbonatedCaffeine | How am I supposed to show my logs if I can't get any | 22:43 |
| T42 | <elros34> on your host/pc | 22:44 |
| CarbonatedCaffeine | I don't know how to do that, I don't route ips | 22:45 |
| CarbonatedCaffeine | or whatever | 22:45 |
| T42 | <edp_17> Don't you have dmesg, journal or whatewer on your host that shows what's happening when you try to connect? | 22:46 |
| CarbonatedCaffeine | No? | 22:46 |
| T42 | <edp_17> Is that a linux system? | 22:47 |
| CarbonatedCaffeine | yes | 22:47 |
| T42 | <edp_17> Then you should have some those I wrote. | 22:47 |
| CarbonatedCaffeine | no | 22:48 |
| T42 | <elros34> what are you talking about run: 'dmesg -w' connect device, run ifconfig commands from hadk-hot->halium instruction ip route/address and again ifconfig | 22:49 |
| CarbonatedCaffeine | ??? | 22:49 |
| T42 | <edp_17> I used this to bring up the usb network, then I was able to telnet to the device: http://docs.halium.org/en/latest/porting/debug-build/early-init.html | 22:51 |
| CarbonatedCaffeine | I tried to do that omg | 22:51 |
| CarbonatedCaffeine | that didn't work | 22:51 |
| T42 | <elros34> and again "didn't work". Are you going to show anything I asked or not? | 22:52 |
| CarbonatedCaffeine | No | 22:52 |
| T42 | <elros34> fair enough. Sorry if I was rough. We can not communicate so somebody else will probable help you better/faster | 22:54 |
| T42 | <edp_17> I had a device where I wasn't able to bring up usb network, chosing a different android base has solved the problem for me. | 22:57 |
| CarbonatedCaffeine | Man, I'm not starting all over again | 22:57 |
| CarbonatedCaffeine | If I have ta do that, then I'm kinda done | 23:05 |
| mal | I don't understand why usb networking would suddenly stop working | 23:16 |
| mal | also there is almost never need to start all over again | 23:16 |
| CarbonatedCaffeine | Hi, sorry I got frustrated. I took the time to read the init docs and I was able to enter the phone again via telnet. | 23:17 |
| CarbonatedCaffeine | What logs can I get | 23:29 |
| CarbonatedCaffeine | thanks | 23:29 |
| mal | journal log of course, if not possible then dmesg | 23:30 |
| CarbonatedCaffeine | https://pastebin.com/zhESTQDj | 23:35 |
| CarbonatedCaffeine | Dmesg is the best I have currently | 23:35 |
| T42 | <Mister_Magister> @b100dian since flashing anything on recend devices is massive pain in the butt i think it will be good idea to change the sfos image to more official one like in fxtec in future | 23:46 |
| T42 | <Mister_Magister> where you just fastboot all the partitions | 23:46 |
| mal | how far does the device boot? | 23:55 |
| mal | if that from before systemd is up and if you continue to boot causes reboot first thing to do is to mask droid-hal-init | 23:56 |
Generated by irclog2html.py 2.17.1 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!