**** BEGIN LOGGING AT Fri Nov 29 02:59:58 2019 Nov 29 03:49:44 Is there a way to port Lineage 14.1 to my device? I only see a device tree for 15.1 + Nov 29 03:57:36 I found an unofficial port of 14.1 for my device, but see no github for it. Any possible way to extract the information such as dependencies, etcetera; for my manifest file? Nov 29 03:58:15 [Edit] I found an unofficial port of 14.1 for my device, but see no github for it. Any possible way to extract the information from the actual .zip file such as dependencies, etcetera; for my manifest file? Nov 29 03:59:16 If you can find the sources then it should work Nov 29 03:59:25 Things like the device tree, kernel source and vendor blobs Nov 29 03:59:45 Searching now! Because I'll report my findings. Nov 29 04:05:30 I've located the developer's github. Extensively searching for my device's source pertaining to his Lineage 14.1 port. Nov 29 04:08:09 No luck. :( Nov 29 04:10:47 Actually. Nov 29 04:11:12 Found it! Nov 29 05:03:31 @TuxThePenguin [Things like the device tree, kernel source and vendor blobs], I can't find my kernel source. Nov 29 06:30:07 @SomeIrrelevantPerson0000 [I found an unofficial port of 14.1 for my device, but see no github for it. Any …], Thats what I dislike about the whole Android culture, even the unofficial ROM porters did not bother to put their sources somewhere, just they publish zip files where nobody knows whats going on. … On the other hand the vendor bl Nov 29 06:30:07 obs are illegal to publish with most vendors, and it can get you in troubles having them. Nov 29 06:37:46 @SomeIrrelevantPerson0000 [I found an unofficial port of 14.1 for my device, but see no github for it. Any …], Everything is as usual. But even if there were source codes for your smartphone, you would hardly be able to assemble the firmware according to the instructions. It is difficult for those who understand this, but for those who Nov 29 06:37:46 do not understand it, it is practically impossible. In my case, nothing happens, because the mount points are different and how to solve the problem is not entirely clear to me. In your case, it may turn out exactly the same. Nov 29 07:20:30 Is there some lunos porting group? Nov 29 08:09:08 @Lukapanio: #webos-ports ;) Nov 29 08:09:32 Thanks Nov 29 08:47:32 Should I be concerned about these lines? https://paste.ubuntu.com/p/npf5gJhgG3/ Nov 29 08:47:49 That's during building. Nov 29 10:08:49 @rixchy [That's during building.], Yes, you need to check that configs in the defconfig and remove the #config ... Is not set Nov 29 10:08:59 @rixchy [That's during building.], [Edit] Yes, you need to check those configs in the defconfig and remove the #config ... Is not set Nov 29 10:10:34 @rixchy [Should I be concerned about these lines? https://paste.ubuntu.com/p/npf5gJhgG3/], This means you have in your defconfig … #CONFIG_WATCHDOG_NOWAYOUT is not set … CONFIG_WATCHDOG_NOWAYOUT=y … So by default this option is always disabled Nov 29 10:10:46 @rixchy [Should I be concerned about these lines? https://paste.ubuntu.com/p/npf5gJhgG3/], [Edit] This means you have in your defconfig … #CONFIG_WATCHDOG_NOWAYOUT is not set … CONFIG_WATCHDOG_NOWAYOUT=y … So this option is always disabled Nov 29 10:11:26 @rixchy [Should I be concerned about these lines? https://paste.ubuntu.com/p/npf5gJhgG3/], [Edit] This means you have in your defconfig … #CONFIG_WATCHDOG_NOWAYOUT is not set … CONFIG_WATCHDOG_NOWAYOUT=(y/n) … So this option is always disabled Nov 29 10:12:02 @rixchy [Should I be concerned about these lines? https://paste.ubuntu.com/p/npf5gJhgG3/], [Edit] This means you have in your defconfig for example: … #CONFIG_WATCHDOG_NOWAYOUT is not set … CONFIG_WATCHDOG_NOWAYOUT=(y/n) … So this option is always disabled Nov 29 10:12:12 @rixchy [Should I be concerned about these lines? https://paste.ubuntu.com/p/npf5gJhgG3/], [Edit] This means you have in your defconfig, for example: … #CONFIG_WATCHDOG_NOWAYOUT is not set … CONFIG_WATCHDOG_NOWAYOUT=(y/n) … So this option is always disabled Nov 29 10:38:58 @Flohack [Thats what I dislike about the whole Android culture, even the unofficial ROM po …], I did notice going through the halium porting guide, that some devices would have similar layouts. That is the case of my device. But however, the kernel source is nowhere to be found. Maybe if anybody wants to take a look and h Nov 29 10:38:59 elp? Nov 29 10:40:56 @Flohack [Thats what I dislike about the whole Android culture, even the unofficial ROM po …], [Edit] I did notice going through the halium porting guide, that some devices would have similar layouts. That is the case of my device. But however, the kernel source is nowhere to be found. Maybe if anybody wants to take a loo Nov 29 10:40:56 k and help? My device I'm working on is the SM-T350. Samsung Galaxy Tab A 8.0 Nov 29 10:50:18 @Flohack [Thats what I dislike about the whole Android culture, even the unofficial ROM po …], [Edit] I did notice going through the halium porting guide, that some devices would have similar layouts. That is the case of my device. But however, the kernel source is nowhere to be found. Maybe if anybody wants to take a loo Nov 29 10:50:18 k and help? My device I'm working on is the SM-T350. Samsung Galaxy Tab A 8.0; codename: gt58wifi. Nov 29 10:51:24 The android_device is known to be as the gte-common. Nov 29 10:51:47 Vendor is obviously Samsung. Nov 29 11:38:07 @SomeIrrelevantPerson0000 [I did notice going through the halium porting guide, that some devices would hav …], You can try to use a different kernel but the mess is: ARM devices in general and mobile phones specifically do not have any kind of "hardware detection" and configuration is mostly static. Means a wrong kernel could not find ce Nov 29 11:38:07 rtain hardware parts, or even misconfigure them. If you think about timing and voltages. In a mobile phone this can be very critical and even lead to damages Nov 29 11:38:32 So a wrong kernel comes with a wrong driver that will output some sh1t to the I2C bus, resulting in a fried regulator or GPU etc Nov 29 11:39:15 @Flohack [So a wrong kernel comes with a wrong driver that will output some sh1t to the I2 …], Reading the sentence "resulting in a fried regulator or GPU" makes me say ouch. Nov 29 11:39:15 An Android kernel on startup "assumes" that all the hardware its made for is in the right place,on the right bus, on the right address. So yeah be careful ^^ Nov 29 11:39:44 can anyone share a log of successful … result of command ./halium-install -p ut ubports-touch.rootfs-xenial-armhf.tar.gz system.img Nov 29 11:39:57 Well yes while I dont know of any such case, its a delicate thing. The whole smartphone is not made to have exchangeable hardware, so its not expected that hardware can ever change. Nov 29 11:40:21 mine ends with I: Cleaning up … umount: .halium-install-rootfs.3KZI8: not mounted … should I expect more Nov 29 11:41:04 @nitin_3 [mine ends with I: Cleaning up … umount: .halium-install-rootfs.3KZI8: not mounted …], No. Check with adb shell whats in the /data folder Nov 29 11:41:07 and tell us Nov 29 11:41:52 @Mattia990 [Yes, you need to check those configs in the defconfig and remove the #config ... …], @rixchy did you read this? Nov 29 11:42:06 I did. Nov 29 11:42:24 @rixchy [I did.], Did you resolve your problems? Nov 29 11:42:42 @nitin_3 [/data folder is same...it didn't get wiped], Well it should not be wiped, but whats inside? make `ls` Nov 29 11:42:45 Not yet, my kernel somehow broke and now it's giving alot of function problems. Nov 29 11:42:49 @Mattia990 [Did you resolve your problems?], . Nov 29 11:43:22 I just didn't care, renamed the old kernel folder, and downloaded another one again. Nov 29 11:43:32 Now, there's a problem, that I know to fix. Nov 29 11:43:37 @rixchy [I just didn't care, renamed the old kernel folder, and downloaded another one ag …], Yeah try in this way Nov 29 11:43:39 [Edit] Now, there's a problem, that I know how to fix. Nov 29 11:44:38 The only thing that I don't know how to fix, is this error. "error: unknown type name '__kernel_fd_set' … typedef __kernel_fd_set fd_set;" Nov 29 11:45:17 @rixchy [The only thing that I don't know how to fix, is this error. "error: unknown type …], I think it is in the ubports pad or in the halium documentation Nov 29 11:45:25 @rixchy [The only thing that I don't know how to fix, is this error. "error: unknown type …], I had this problem Nov 29 11:45:51 This error too? "/home/rixchy/halium/kernel/samsung/exynos5420/security/apparmor/domain.c:783:13: error: 'struct task_struct' has no member named 'no_new_privs' … if (current->no_new_privs && !unconfined(profile)) {" You had this error too, perhaps? Nov 29 11:46:03 @Flohack [Well it should not be wiped, but whats inside? make ls], https://pastebin.com/mLEM7NEk Nov 29 11:47:08 I have stuff that should be in /data folder plus rootfs and also system.img Nov 29 11:49:19 @Mattia990 [I had this problem], Do you remember where you got the patches or something? I can't seem to find the fixes on halium or on ubports. Nov 29 11:52:06 @Flohack [An Android kernel on startup "assumes" that all the hardware its made for is in …], So in other words, no hope for me? Lol Nov 29 11:54:13 @nitin_3 [I have stuff that should be in /data folder plus rootfs and also system.img], Wow thats a lot. For UT you can delete all stuff in this folder and then re-run halium install. It should leave rootfs.img and system.img not more Nov 29 11:54:39 Or use TWRP to delete it that might be easier Nov 29 11:54:55 yeah wiped data partition Nov 29 11:56:25 i am able to telnet into device at port 23 ..is it bad or good? Nov 29 11:56:58 @nitin_3 [i am able to telnet into device at port 23 ..is it bad or good?], Hmm it basically means your halium boot could not find the rootfs probably Nov 29 11:57:06 So kernel is alive but thats it Nov 29 11:58:15 what could be causing this.. Nov 29 12:08:40 "error: 'LSM_UNSAFE_NO_NEW_PRIVS' undeclared" I used `grep` and it indeed isn't declared anywhere.. Nov 29 12:13:02 @Flohack [Hmm it basically means your halium boot could not find the rootfs probably], [ 5.531405] initrd: checking filesystem integrity for the userdata partition … [ 5.922055] EXT4-fs (sda21): mounted filesystem with ordered data mode. Opts: errors=remount-ro … could this be causing the issue Nov 29 12:15:35 @rixchy [The only thing that I don't know how to fix, is this error. "error: unknown type …], Maybe this? … https://github.com/thestr4ng3r/chiaki/issues/49 Nov 29 12:19:26 Nope, didn't work. Nov 29 12:19:44 No `sys/select.h` is found. Nov 29 13:13:41 @nitin_3 [[ 5.531405] initrd: checking filesystem integrity for the userdata partition …], No, thats not an issue Nov 29 13:14:10 @rixchy ["error: 'LSM_UNSAFE_NO_NEW_PRIVS' undeclared" I used grep and it indeed isn't de …], I think thats in Halium docs somewhere how to fix that Nov 29 13:15:11 @Flohack [No, thats not an issue], would dmesg from port 23 telnet help?? Nov 29 13:20:50 Yes maybe Nov 29 14:02:15 hello ,ecnourted the same problem as @nitin_3 while install ubuntu rootfs here are my logs Nov 29 14:02:23 https://pastebin.com/WSiLnupC Nov 29 14:02:37 and the halium-install script output Nov 29 14:02:38 https://pastebin.com/BLbr2nMq Nov 29 14:03:07 wiped /data partition and reinstalled using the script only system.img and rootfs.img Nov 29 14:04:51 using telnet and then using netstat -a gives only telnet port i think the ssh service isn't running if anyone can direct me to the problem / how to solve thanks Nov 29 14:08:42 @Flohack [Yes maybe], here is dmesg from port 23 telnet … https://pastebin.com/EADtfpt4 Nov 29 14:12:34 @nitin_3 [here is dmesg from port 23 telnet … https://pastebin.com/EADtfpt4], `[ 7.895548] run-init: opening console: No such device` try removing console= setting from kernel cmdline Nov 29 14:12:50 Same for @Qusay362 maybe Nov 29 14:14:10 Thats also weird: ```[ 7.693790] initrd: checking fstab /root/var/lib/lxc/android/rootfs/fstab* for additional mount points … [ 7.695759] cat: can't open '/root/var/lib/lxc/android/rootfs/fstab*': No such file or directory``` Nov 29 14:18:17 @Flohack [Thats also weird: [ 7.693790] initrd: checking fstab /root/var/lib/lxc/androi …], I do have fstab.qcom file … but it is inside vendor/etc/ Nov 29 14:40:11 @nitin_3 [I do have fstab.qcom file … but it is inside vendor/etc/], Can you check in the $OUT directory right after you bilt the systemimage? Nov 29 14:42:02 @Flohack [Can you check in the $OUT directory right after you bilt the systemimage?], ok...I am building again..so It would take some time Nov 29 14:44:30 You should find the file there in system or root folder? dunno now, at my workplace Nov 29 14:45:35 But definitely the fstab should be copied into the right folder, and then it will appear visible to the lxc. But, to be honest, that second step, first fix console issue. For that you need to rebuild kernel and systemimage, and install both. There is the replace-android-image script to not having to install also rootfs every time Nov 29 14:55:45 there are some configs..that conflict with each other for halium and hybris boot image like merkernel check says enable CONFIG_BT_HCIUART … whereas halium kernel check unsets it Nov 29 15:02:03 yes true Nov 29 15:02:12 Thats the fate of 2 different operating systems Nov 29 15:02:20 Its not about Halium boot at all but for later stages Nov 29 15:02:32 CONFIG_ANDROID_LOW_MEMORY_KILLER … CONFIG_NET_CLS_CGROUP … CONFIG_NETPRIO_CGROUP … CONFIG_BT_HCIUART … these 4 I found Nov 29 15:02:37 Depends on what will be your final case Nov 29 15:02:57 If you want to build for UT, then use UT checker. If you want to go for Plasma Mobile, use mer checker Nov 29 15:03:36 I am trying to build ut so I used halium one Nov 29 15:15:51 Ok thats good Nov 29 16:06:59 @intel777 [but now lxc won't start because it can't find init.rc of android … Feb 11 16:28:02 …], Can I just copy this scripts from one that system.img contains? Nov 29 16:36:10 I have an issue with which no one has helped me with yet. https://pastebin.com/JNcSQp4J Nov 29 16:41:02 Oh, this is the types Nov 29 16:41:13 undeclared stuff again Nov 29 16:41:17 I guess Nov 29 16:41:22 Yep. But.. undeclared? Nov 29 16:42:00 yeah, it's definitely not C types, so they must be described somewhere Nov 29 16:42:29 Okay. Nov 29 16:42:40 Interesting that this is in fact place where they declares Nov 29 16:42:57 typedef Nov 29 16:43:04 Wait. Nov 29 16:43:28 I used `grep __kernel_ino_t ` in my device's and in hammerhead's kernel folder.. Nov 29 16:44:54 @intel777 [typedef], The typedef is an advance feature in C language which allows us to create an alias or new name for an existing type or user defined type. The syntax of typedef is as follows: Syntax: typedef data_type new_name; typedef : It is a keyword Nov 29 16:45:42 So it needs to be created like: "typedef unsigned __kernel_fd_set" Nov 29 16:45:50 But there's a weird thing. Nov 29 16:45:54 And a catch. Nov 29 16:46:40 It isn't declared neither in my device's kernel, neither in hammerhead's, so I think that file does not have a problem. Nov 29 16:59:40 I think it's `typedef`'d in `include/linux/posix_types.h`. Nov 29 16:59:47 `typedef struct { … unsigned long fds_bits [__FDSET_LONGS]; … } __kernel_fd_set;` Nov 29 17:05:40 root folder https://pastebin.com/jU0m3cD7 … system folder https://del.dog/uwiwaipput.txt … and vendor folder https://del.dog/esotofeqyp.txt Nov 29 17:06:23 @Flohack [You should find the file there in system or root folder? dunno now, at my workpl …], the fstab.qcom is going to vendor/etc instead of system or root Nov 29 17:09:48 @nitin_3 [the fstab.qcom is going to vendor/etc instead of system or root], That should be fixed by Halium actually but I will check. Nov 29 17:12:38 (Photo, 960x1280) https://irc.ubports.com/xbcsP99r.png Nov 29 17:12:39 (Photo, 1280x960) https://irc.ubports.com/0wsCglzj.png Nov 29 17:13:35 So its working but cant start unity sesdion Nov 29 17:13:48 starting lightdm gives Failed to use bus name org.freedesktop.DisplayManager, do you have appropriate permissions? Nov 29 17:17:50 @Flohack [That should be fixed by Halium actually but I will check.], Do I need to manually copy it? … if yes … where system or root Nov 29 17:25:34 @Qusay362 [starting lightdm gives Failed to use bus name org.freedesktop.DisplayManager, do …], Did you add udev rules Nov 29 17:26:15 It makes no sense to continue until test_hwcomposer works Nov 29 17:26:30 Switch lightdm to disabled until you get everything sone Nov 29 17:29:41 adding this should copy fstab.qcom file … but should I copy it to root/ or system/ … PRODUCT_COPY_FILES += \ … $(LOCAL_PATH)/rootdir/etc/fstab.qcom:root/fstab.qcom Nov 29 17:31:32 @Qusay362 [], What is device name? Nov 29 17:33:31 samsung note 3 codename hltechn Nov 29 17:37:11 @Qusay362 [samsung note 3 codename hltechn], Have Got device problem? Nov 29 17:40:29 ```Feb 11 16:28:32 localhost.localdomain systemd[1]: Started LXC Container: android. … Feb 11 16:28:33 localhost.localdomain lxc-start[1344]: 8082 blocks … Feb 11 16:28:33 localhost.localdomain lxc-start[1344]: lxc-start: conf.c: remount_all_slave: 3820 Invalid argument - Fai … Feb 11 16:28:33 localhost.localdomain lxc-start[1344]: lxc-s Nov 29 17:40:29 tart: conf.c: remount_all_slave: 3821 Continuing...``` … Is that means that LXC started successfully? Nov 29 17:40:58 oh, I see fail... nevermind Nov 29 17:41:34 ```conf.c: remount_all_slave: 3820 Invalid argument - Failed to make /var/lib/lxc/android/rootfs/socket rslave``` Nov 29 17:42:36 @intel777 [Feb 11 16:28:32 localhost.localdomain systemd[1]: Started LXC Container: android …], Did you check lyc flags with lxc-checkconfig? Nov 29 17:43:11 Yeah, I modified kernel configs to met lxc-checkconfig requirements Nov 29 17:43:45 literally every check is 'enabled' Nov 29 17:44:33 on, I see now … ```newuidmap is not installed … newgidmap is not installed``` Nov 29 17:44:37 @rixchy [I have an issue with which no one has helped me with yet. https://pastebin.com/J …], So, anyone knows how to fix this.....? Nov 29 17:44:47 @intel777 [on, I see now … newuidmap is not installed … newgidmap is not installed], its namespaces problem Nov 29 17:46:48 @intel777 [on, I see now … newuidmap is not installed … newgidmap is not installed], well, I can't find this in kernel config Nov 29 17:47:15 oh, it's just packages Nov 29 17:52:08 oh, how can I transfer file to device? Nov 29 17:52:54 sftp desn't respond. ftp, scp is not there Nov 29 17:53:25 as well as wget and curl Nov 29 18:00:25 oh, I feel like magician now … ``` … cat filename | ssh user@host "cat > remote" Nov 29 18:00:27 [Edit] oh, I feel like magician now … ```cat filename | ssh user@host "cat > remote"``` Nov 29 18:30:22 @rixchy [So, anyone knows how to fix this.....?], So, I kind of found a solution online. It says, that you need to include `#include ` in the `types.h` file. But where is that select.h file? I cannot seem to find it! Nov 29 18:58:49 M M was added by: M M Nov 29 19:22:05 @rixchy [So, I kind of found a solution online. It says, that you need to include #includ …], look at your log carefully...it has path to types.h there Nov 29 19:22:16 @nitin_3 [look at your log carefully...it has path to types.h there], I know. Nov 29 19:44:30 @rixchy [So, I kind of found a solution online. It says, that you need to include #includ …], Another broken Android kernel. They mess them up pretty well Nov 29 19:44:58 AaAaAa. I have been trying for day now. Nov 29 19:45:10 Applying fixes, didn't work, removing. Nov 29 19:46:50 Maybe its just too broken to be modified. Android kernels are zombies sometimes Nov 29 19:47:47 Well, I have seen alot of threads that say the fix, but it doesn't work. Plus, @Mattia990 encountered the same issue, tho he didn't remember what did he apply. Nov 29 19:47:56 [Edit] Well, I have seen alot of threads that say the fix, but it doesn't work. Plus, Mattia encountered the same issue, tho he didn't remember what did he apply. Nov 29 19:48:26 And he seemed to fix it, so I don't really know. Nov 29 19:49:05 @Mattia990 [Maybe this? … https://github.com/thestr4ng3r/chiaki/issues/49], He suggested this. Nov 29 20:03:19 now not getting telnet Nov 29 21:27:50 Ok, so. I downloaded another kernel from another repo, I didn't expect it to work, but I would try to fix it, to see if it has the same errors. So, I applied somethings, aand, now I got this.. https://pastebin.com/9HhKzTFc The fact that it's at the beginning of the build... Nov 30 02:08:41 Silvia Black was added by: Silvia Black **** ENDING LOGGING AT Sat Nov 30 02:59:57 2019