**** BEGIN LOGGING AT Fri Feb 09 03:00:02 2018 Feb 09 03:00:38 rockybulwinkle was added by: rockybulwinkle Feb 09 03:04:34 I'm rockybulwinkle from before. Decided telegram would be more reliable than irc. Feb 09 03:22:53 it seems doze was bringing in SettingsLib. I removed it from the repo and it seems to be better now? I assume since that's an android-specific app we don't need it. Feb 09 03:24:00 yeah LOS doze packages are only for android systemui stuff Feb 09 03:26:31 should I be worried about this message when I breakfast? "Trying dependencies-only mode on a non-existing device tree?" Do I need to define my device tree/find where LOS defines it? Feb 09 03:31:58 I don't think so. Feb 09 03:32:03 I get that error when I build. Feb 09 03:32:47 cool Feb 09 03:33:04 I think it's just the build system complaining. Feb 09 03:33:47 Next problem.. it can't find signapk.jar? https://gist.github.com/rockybulwinkle/c6ac470561b66fd1e217db6c481d7df4 … I saw it in prebuilts so I just copied it to where it was expecting it. Is that what I'm meant to do? Feb 09 03:34:46 go to your vendor folder for the device and figure out what's calling the apks and comment it out Feb 09 03:34:56 We don't need them for halium Feb 09 03:35:44 haha ok. thanks. Feb 09 03:37:58 ok, I added a new repo to my manifest, android_device_qcom_sepolicy, I hope I can use breakfast after that Feb 09 03:43:55 It's complaining it can't find libcryptfs_hw_intermediates now when building systemimage. This is where I got stuck last night... https://gist.github.com/rockybulwinkle/c577b831bee2a980b462e50cebf52441 Feb 09 03:44:59 I had tried adding to my board config: TW_INCLUDE_CRYPTO := true … TARGET_HW_DISK_ENCRYPTION := true … TARGET_CRYPTFS_HW_PATH := device/qcom/common/cryptpfs_hw Feb 09 03:45:32 Hm, it's missing the actual library Feb 09 03:45:35 Is it in your vendor? Feb 09 03:50:07 hmm I coulda sworn it was at that path yesterday. I gotta do some greping Feb 09 03:50:09 https://github.com/LineageOS/android_vendor_qcom_opensource_cryptfs_hw Feb 09 03:56:17 Awesome! Now it complains about external/ant-wireless/antradio-library/com.dsi.ant.antradio_library.xml. Is that something from android or something we need for halium? Feb 09 03:58:32 https://github.com/LineageOS/android_external_ant-wireless_antradio-library Feb 09 03:58:58 I'd say its safe enough to comment out, its only for connecting your device to certain fitness trackers that use proprietary software anyway Feb 09 03:59:16 ah. Feb 09 04:00:33 Hey! It gets to starting the build! Feb 09 04:01:22 Don't worry, I'm sure there'll be imminent disappointment. :P Feb 09 04:02:38 I do have some questions about the kernel config. I didn't see the config where the porting guide said it would be. I assume I can just use the default config for this CPU with the modifications mentioned in the guide? Feb 09 04:02:53 I think that's what LOS was doing anyway. build's at 10% Feb 09 04:03:13 Uh Feb 09 04:03:36 Check your device's BoardConfig Feb 09 04:03:55 lineageos_axon7_defconfig Feb 09 04:04:33 I think the axon7 is aarch64 so it might be under a different folder in arch/configs Feb 09 04:05:25 I see now. Feb 09 04:05:46 Plus they split their boardconfig into multiple include's Feb 09 04:06:00 yeah I saw that when I looked up the device tree...what the hell x3 Feb 09 04:07:33 when I get to the point that I flash halium to my phone, it doesn't overwrite TWRP does it? Feb 09 04:07:57 Uh, it shouldn't Feb 09 04:08:08 Unless I make a mistake :p Feb 09 04:08:11 Yeah. Feb 09 04:08:27 Make sure there's nothing on your phone you care about before flashing anything. Feb 09 04:09:02 I was already planning to make a backup Feb 09 04:16:56 @Lyokanthrope, Oh, how did you know? It couldn't find the cryptfs_hw.h Feb 09 04:21:28 Uhm, try explicitly building it first if you've added it to your manifest and sync'd Feb 09 04:21:38 I forget the command to do that...mmm or something @UniversalSuperBox Feb 09 04:24:28 I added the path for cryptfs_hw to common_c_includes in system/vold/Android.mk and it seems to be building now Feb 09 04:32:25 mka systemimage completed! Feb 09 04:33:33 Huzzah! Feb 09 04:35:23 Now I just need to fix the kernel configuration. Looks like a lot of values are wrong. Feb 09 04:35:32 Thanks for the help Lyo, I really appreciate it! Feb 09 04:36:12 You're welcome! I need to try and be more proactive about helping in here, lol Feb 09 04:40:39 Uhhhh hm. So menuconfig says this about LBDAF: https://gist.github.com/rockybulwinkle/c09f777dae7bd092ee8ca0d5fba76ece Feb 09 04:41:05 It depends on !64bit... but I'm a 64 bit device... and that 'mer' tool wants me to set it to yes? I'm confuse. Feb 09 04:41:32 I don't think the mer kernel check takes aarch64 into account Feb 09 04:42:04 ok. I'm gonna guess that 64 bit linux can handle ext4 huge_file without extra config? Feb 09 04:42:15 I'm...honestly not sure. Feb 09 04:44:14 wow. So that lets files be up to 2TB.... My phone is only 64GB I think? Maybe we should just make our ext4 w/o huge_file support. Feb 09 04:49:50 are we using systemd or some sysvinit system? Feb 09 04:51:53 I think systemd, though ubports still uses upstart to my knowledge Feb 09 04:51:57 I ask because the mer kernel check warns about some options and mentions systemd. Feb 09 04:52:22 Yeah, you'll want to enable all the systemd stuff and anything related to cgroups for sure Feb 09 04:55:16 I'm not seeing the options for stuff whose comment says, "systemd (optional): ht … tp://0pointer.de/blog/projects/cgrou … ps-vs-cgroups.html" Feb 09 04:55:54 the website doesn't give a patch or anything.... Feb 09 04:58:25 Don't worry about it. Feb 09 04:58:47 @rockybulwinkle, What's the option for it? Feb 09 04:59:01 Lots of things, I'll make a list Feb 09 05:00:29 mer kernel config also has some contradicting options, for example it wants SECURITY_SELINUX_BOOTPARAM but that requires selinux which requires AUDIT, but mer wants AUDIT to be 'n'. Feb 09 05:01:54 @rockybulwinkle, You can disable audit in board Feb 09 05:03:06 https://gist.github.com/rockybulwinkle/4b91337b874c1af866ce8d5aad1a2fa3 Feb 09 05:03:39 Enable all of them. Feb 09 05:04:39 Some of them affect lxc-checkconfig status Feb 09 05:05:30 and you should be carefu as I see one entry repeated twice. Feb 09 05:05:50 They aren't present in my Kconfigs Feb 09 05:06:09 Kernel version? Feb 09 05:06:14 3.18 Feb 09 05:06:30 I think it said it needed to be more than 3.1? Feb 09 05:07:06 Just enable those stuff in defconfig and try building. If it gives error we wil see abut that. Feb 09 05:07:31 ok Feb 09 05:08:00 I need to head to bed though. Thanks for all the help! Feb 09 05:36:02 Anybody knows a good tool to compare kernel configs? I knew I used something in the past but I can't find it anymore Feb 09 05:41:03 nvm got it : halium/kernel/google/msm/scripts/diffconfig Feb 09 06:25:47 Link to project planning on bottom of https://halium.org/announcements/2017/04/24/halium-is-in-the-air.html should be https://github.com/Halium/docs/blob/master/Planning.rst Feb 09 06:31:36 Found it https://github.com/Halium/halium.org/pull/24 Feb 09 06:32:02 merged Feb 09 06:33:05 :) Feb 09 07:44:20 @rockybulwinkle, You can just remove the line that includes settingslib in your device tree. SettingsLib is most likely Java, and that's not used in halium Feb 09 07:46:57 Sorry, I think that was already answered :) I'm just reading the back log Feb 09 10:59:42 Hi, I'm the Le Pro 3 guy, after downloading sepolicy repo I get "Trying dependencies-only mode on a non-existing device tree?" what should I do? thanks for your help Feb 09 11:02:35 @Pablo Alejandro, sounds like your lunch or breakfast command wsa not successful Feb 09 11:07:51 Lol, obvioulsy ir was not successful, but thanks Feb 09 11:09:03 Before downloading that repo I got "don't have a product spec for cm_zl1" or something like that Feb 09 11:09:26 the error you pasted is not full error Feb 09 11:09:33 can you pastebin the full log? Feb 09 11:16:47 @Pablo Alejandro, You probably have the issue that the build scripts dont detect all files in the device and vendort tree. Sometimes you must create these files yourself. Are you using Cm or LineageOS sources? Feb 09 11:31:24 LineageOS sources, sorry for no posting full error, I'm in my school right now, when I arrive at my house I will send it Feb 09 13:39:36 @Pablo Alejandro, I was getting the same error, should I be concerned? Feb 09 13:41:16 I did manage to make a system image having that error but I'm concerned that it didn't find _any_ device tree Feb 09 13:42:30 I haven't tested my system image yet. Feb 09 13:43:45 validovic was added by: validovic Feb 09 14:00:35 KunalBhat was added by: KunalBhat Feb 09 14:00:40 Hello Feb 09 14:00:59 Fwd from KunalBhat: Could I get a little help here ? Feb 09 14:01:00 Fwd from KunalBhat: ninja: Entering directory `.' … ninja: error: '/home/kunal_bhat/helium/out/host/linux-x86/framework/signapk.jar', needed by '/home/kunal_bhat/helium/out/target/product/z3c/obj/APPS/TimeService_intermediates/package.apk', missing and no known rule to make it … build/core/ninja.mk:151: recipe for target 'ninja_wrapper' failed … make: *** [ninja_wrapper] Error 1 … make: Leaving directory '/home/kunal_bhat/helium' … #### Feb 09 14:01:01 to build some targets (5 seconds) #### Feb 09 14:01:02 Fwd from KunalBhat: I get this error when trying to build the Halium systemimage Feb 09 14:01:03 Fwd from KunalBhat: could someone tell me where I'm going wrong ? Thanks Feb 09 14:02:27 That looks like its pulling in an android package. Find where that is happening and comment it out. Feb 09 14:03:44 Alright Feb 09 14:03:45 thanks Feb 09 14:38:56 So i think I forgot to fixup mountpoints Feb 09 14:40:30 But when I'm typing "readlink -f [my source path]" there comes no output. I'm reinstalling my OS on my second PC to test it if its working on the second device Feb 09 14:40:35 Fwd from KunalBhat: ninja: error: 'external/ant-wireless/antradio-library/com.dsi.ant.antradio_library.xml', needed by '/home/kunal_bhat/helium/out/target/product/z3c/system/etc/permissions/com.dsi.ant.antradio_library.xml', missing and no known rule to make it Feb 09 14:42:01 @KunalBhat that can probably be commented out too. Someone else told me it was for fitness trackers. Feb 09 14:42:18 Sorry I'm not more specific, I don't have the source in front of me right now Feb 09 14:43:08 Not an issue Feb 09 14:43:09 @Manu, Could this be the reason why the mka command didn't work? Feb 09 14:43:17 I'm trying to find the file rn Feb 09 14:43:21 ok Feb 09 14:44:07 @Manu, You are executing the commands on your android device, correct? Feb 09 14:44:17 No Feb 09 14:44:36 where are you running them? Feb 09 14:44:37 On my Linux machine Feb 09 14:44:47 you need to connect over adb to your device and log in. Feb 09 14:45:10 Okay what is meant with login? Feb 09 14:45:29 http://docs.halium.org/en/latest/porting/build-sources.html#include-your-device-in-fixup-mountpoints Feb 09 14:45:51 ADB let's you connect to your device and get a terminal running /on the device/ Feb 09 14:46:43 Ah okay thanks Feb 09 14:47:14 You should be able to find a tutorial :) Feb 09 14:47:26 Okay Feb 09 14:57:17 @KunalBhat, In /vendor/your/device/Android.mk … Remove all code with .apk our .jar in Feb 09 14:57:56 Yeah i fixed that Feb 09 14:58:20 I'm trying to find antradio library Feb 09 15:06:18 this is where I found mine: ./device/zte/axon7/product/ant.mk Feb 09 16:09:38 bhash: ok Feb 09 16:09:55 @NotKit: does /system/bin/direct_sensors_test or something like that work? Feb 09 16:10:31 need to build it Feb 09 16:10:44 does Ubuntu use compat libs for sensors though? Feb 09 16:11:29 I think it does Feb 09 16:13:49 there is only direct_input_test Feb 09 16:13:55 from compat/input Feb 09 16:14:15 mmm binary maybe calleed something else.. but there is something for sure Feb 09 16:33:11 I checked compat/input code, nothing that could be sensor-related there Feb 09 16:36:05 I guess Mir nowadays uses libinput, since I don't have that library at all and input works Feb 09 16:37:04 I think : https://github.com/ubports/platform-api here is something sensor related Feb 09 16:45:46 I think I found a solution for the problem that we need to change the vendor repositories but don't want to fork them. Do you think this would solve it? … https://github.com/Halium/halium-devices/pull/38 Feb 09 17:54:45 I've been getting odd errors while building the linux kernel for zte axon7.. I was able to resolve most of them but this one is stumping me. Seems I'm missing dma.h for my chip??? /mnt/foo/halium-dev/kernel/zte/msm8996/drivers/crypto/msm/qce.c:31:22: fatal error: mach/dma.h: No such file or directory Feb 09 17:56:41 https://gist.github.com/rockybulwinkle/f2373074daa9820f91166abbd83885ca Feb 09 17:58:31 this is the kernel I'm using, branch cm14.1 https://github.com/rockybulwinkle/android_kernel_zte_msm8996 Feb 09 18:05:18 does arm64 use arm's dma.h, not an arm64 version? Feb 09 18:09:14 I try using arm's (32 bit) dma.h... the compilation continues... I saw mainline doesn't even have an arch/arm64/include/asm/mach/dma.h so I am assuming it uses the 32 bit arm one. Feb 09 18:18:23 nvm I found the headers... I'm not sure why they weren't being included in the repo: https://android.googlesource.com/kernel/msm.git/+/42bad328ba45ee4fe93216e7e99fe79a782d9155/arch/arm/mach-msm/include/mach Feb 09 20:11:34 So, for whatever reason my ARCH variable wasn't set while doing the menuconfig and I was configuring for x86, which was the sort of a lot of trouble I was having. Do one of the halium scripts set the ARCH env variable? Feb 09 20:11:58 @rockybulwinkle, nop, it was not set for me as well Feb 09 20:12:13 you were supposed to set it when doing menuconfig, I think Feb 09 20:12:27 when building kernel it gets set by Android build system Feb 09 20:12:46 you might also need to specify a path to your defconfig manually Feb 09 20:13:26 I did "make lineageos_axon7_defconfig" to load it into the .config before doing all the menuconfig stuff. I may have just tripped up at some point while modifying it. Feb 09 20:13:30 nbd Feb 09 20:47:35 'undefined reference to `__cpu_method_of_table' ... any ideas? https://gist.github.com/rockybulwinkle/d2d97b061d6928be3df836e6012af4bc Feb 09 20:56:25 There's a second problem I've been running into, when I do a test build the .config disappears. I was assuming it would stay put? Feb 09 21:05:19 @bhushanshah oops Feb 09 21:05:31 is it me or Halium init is missing property service patch? Feb 09 21:06:12 https://github.com/mer-hybris/android_system_core/commit/6d7306445fd6af3a98bb16963561c77a6446f364 Feb 09 21:06:14 this one Feb 09 21:11:23 it's used for getting properties by libhybris, otherwise it tries to read /system/build.prop, which doesn't work for runtime ones Feb 09 21:13:14 interestingly, bionic patch needed for it is there though: https://github.com/Halium/android_bionic/commit/354a90171699ed723dc5fb7f53114e8a5fc7f05b Feb 09 21:13:33 init: property_set("ro.boot.hardware", "qcom") failed ? any idea !!! Feb 09 21:13:40 hey, once I realized mka was pulling .config out from under me and that my ARCH variable wasn't set while doing menuconfig, it actually compiled! Feb 09 21:14:25 @Walid, have it been set before? Feb 09 21:16:28 AHA! I finally figured it out! Feb 09 21:16:49 For some reason, the Android ramdisk has a link from `/vendor` to `/system/vendor` Feb 09 21:16:59 On the Nexus 5X, which absolutely does not have /system/vendor. Feb 09 21:18:09 Now, then... why is that there. Feb 09 21:19:59 maybe it gets created by build system, then removed by init.rc on normal device? Feb 09 21:20:56 Hmm... I don't see anything in the initscripts Feb 09 21:22:00 `BOARD_NEEDS_VENDORIMAGE_SYMLINK := true` Feb 09 21:22:03 methinks I found it Feb 09 21:22:22 I wonder why is it se t for Nexus 5X though Feb 09 21:22:33 time for some git blaming Feb 09 21:23:10 https://github.com/Halium/android_device_lge_bullhead/commit/3673ce2b99034ec5b6a57a129cb050c6121a6ea6 Feb 09 21:23:14 Truly masterful commit messages Feb 09 21:23:37 Unless that makes the `/system/vendor` symlink Feb 09 21:24:32 is it possible that they don't use /vendor partition and provide vendor files inside /system/vendor? Feb 09 21:24:58 Possibly, but then why create a symlink loop? Feb 09 21:25:10 Because there's one at `/system/vendor` that points to `/vendor` Feb 09 21:26:19 misunderstood that Feb 09 21:26:26 so BOARD_NEEDS_VENDORIMAGE_SYMLINK creates /system/vendor symlink Feb 09 21:26:33 Well darn Feb 09 21:27:04 Yeah, link is still in the ramdisk's out Feb 09 21:28:02 @NotKit, in halium-7.1 some properperties passed with cmdline not well set, i will add it in system.prop, my lxccontainer start but not work , Feb 09 21:28:46 https://github.com/Halium/android_system_core/commit/d0339fad007ad5dc01f454a9f8bbb2f6ce43e45d - do you have this commit? Feb 09 21:28:53 Maybe I should change the pre-start script to just delete everything that's in the way Feb 09 21:30:45 I wonder then, does Android's `mount` destructively mount if there's something in the way? Feb 09 21:35:29 in halium-7.1 there is mny pb in permission also is it normal initctl status apparmor give me stop/waiting ? Feb 09 21:42:36 Kaffeine was added by: Kaffeine Feb 09 22:23:29 https://github.com/Halium/android_device_lge_bullhead/blob/halium-7.1/init.bullhead.qseecomd.sh Feb 09 22:23:36 I don't really understand the purpose of this file... Feb 09 22:24:12 I just sits around waiting for keymaster to start, I think. But for some reason my device doesn't have the keymaster binary. Feb 09 22:25:32 It should be on `/firmware`, but it's just not there. Maybe it gets baleeted when the bootloader is unlocked? I'm just spitballing at this point. Feb 09 22:26:18 I am trying to set tethering up and wifi on my device but there is something in container that went wrong: … `root@localhost:~# sudo echo 1 > /dev/wcnss_wlan` … sudo: unable to resolve host localhost.localdomain: Connection refused … echo: write error: Bad address … `root@localhost:~# sudo echo sta > /sys/module/wlan/parameters/fwpath` … sudo: unable to resolve host localhost.localdomain: Connection refused Feb 09 22:26:32 any suggestion, please? Feb 09 22:27:54 root@localhost:~# hostnamectl … Static hostname: localhost.localdomain … Transient hostname: guf-uplus … Icon name: computer … Machine ID: 566e1aaa7591461ca8bebc84cb870e3f … Boot ID: da0df7a584594d968e3a66a8152b39f2 … Operating System: Ubuntu 16.04.2 LTS … Kernel: Linux 3.18.31-perf-gf02d4ab93e6-dirty … Architecture: arm Feb 09 22:28:10 Please read the pin. Feb 09 22:28:40 As for that error, we'll need the output of `dmesg` and `logcat` right after you run those commands to enable the hardware. Feb 09 22:30:22 https://pastebin.com/Kmq4eevk Feb 09 22:30:29 this was dmesg Feb 09 22:34:30 https://pastebin.com/LSsykJis Feb 09 22:52:28 Oh my, that is an unhappy Android Feb 09 22:53:01 I dunno why it magically activated wifi Feb 09 22:53:08 Sometimes it happens late. Feb 09 22:53:09 `ls /var/lib/lxc/android/rootfs/firmware` Feb 09 22:53:47 root@localhost:~# ls /var/lib/lxc/android/rootfs/firmware … total 84 … dr-xr-x--- 6 system system 16384 Jan 1 00:00 ./ … drwxr-xr-x 17 root root 4096 Feb 11 2016 ../ … dr-xr-x--- 4 system system 16384 Feb 3 2018 .Spotlight-V100/ … dr-xr-x--- 2 system system 16384 Feb 3 2018 .fseventsd/ … dr-xr-x--- 3 system system 16384 Jul 5 2011 image/ … dr-xr-x--- 2 system system 16384 Jul 5 2011 verinfo/ Feb 09 22:54:24 you mean that before /firmware was not populated at all Feb 09 22:55:17 It's possible, but unlikely. The Moto G5 Plus had that problem too, starting the wi-fi really late Feb 09 22:55:39 it seems like I am not the only one Feb 09 22:56:49 at this point next question: why isn't it allowing me to ping 8.8.8.8 Feb 09 22:56:58 What's the error Feb 09 22:57:27 `root@localhost:~# ping 8.8.8.8 … socket: Permission denied` Feb 09 22:59:23 set `CONFIG_ANDROID_PARANOID_NETWORK=n` in your defconfig Feb 09 23:01:43 Oh hey, ril started. Feb 09 23:01:51 That's neat. Feb 09 23:26:38 ok, re-flashed hybris-boot and now it is working Feb 09 23:27:20 I am working with a halium rootfs, is it ok to dist-upgrade it? Feb 09 23:29:06 Sure Feb 09 23:29:14 Might run out of space though Feb 09 23:30:11 even though it would like to upgrade lxc-android Feb 09 23:31:28 ok thnx Feb 09 23:32:45 shoud it be as having a xenial? maybe without all the features of UT Feb 09 23:33:49 Yes. Feb 09 23:34:07 You'll have almost none of the features of UT Feb 10 00:01:12 is it possible to download a "generic" UT rootfs to test with halium? or shall I build it? Feb 10 00:01:47 generic means not tailor-made for a specific device Feb 10 00:39:17 Sure Feb 10 00:40:21 ok, I just found it in ubports docs (sorry for stupid question) Feb 10 00:40:22 Trying to load our ci... Feb 10 00:40:34 tht's just it Feb 10 00:41:20 Okay, good since my connection is poor Feb 10 00:42:17 once I will install it, shall I need to do any configuration or to expect it to run as far as it can? Feb 10 00:42:49 Configs are in the docs Feb 10 02:02:49 Where can an arch user get make_ext4? I don't see it in AUR. Feb 10 02:04:47 I pulled it from here: https://github.com/superr/make_ext4fs.git Feb 10 02:12:49 I'm running into a problem when I try to install the system.img and rootfs: https://gist.github.com/rockybulwinkle/4275dce2e808957e06b8e59da60c8e77 Feb 10 02:13:10 "error: ext4_allocate_best_fit_partial: failed to allocate 749 blocks, out of space?" Feb 10 02:14:13 The image is too small for the stuff that's getting packed into it Feb 10 02:14:14 you'll have to modify the install script to make your system image bigger I think Feb 10 02:14:25 oh ok Feb 10 02:15:36 I thought it was talking about out of space on the device somehow Feb 10 02:19:54 Getting further after upping it to a GB Feb 10 02:23:14 Oh boy, first boot..... Feb 10 02:24:11 lol, nothing. Feb 10 02:25:17 So, before I rebooted TWRP claimed there wasn't an OS installed.... Feb 10 02:27:33 It's not installed like a usual Android, so it's possible that TWRP doesn't detect it Feb 10 02:28:15 @rockybulwinkle, nothing in dmesg? Feb 10 02:28:43 @JBBgameich, Yeah that's what I was thinking. Feb 10 02:29:32 @Lyokanthrope, I mean blank screen (well, tux was there for a bit) and then it rebooted. Looking at sys/fs/pstore/console-ramoops now. So it really stores the last log in _ram_? It looks pretty corrupted, I assume the ram wasn't being refreshed. Feb 10 02:29:36 is there any better way? Feb 10 02:32:20 @rockybulwinkle, That's the only place... At least it's mainlined now I guess Feb 10 02:33:47 So @vanyasem I (but actually Marius) figured out the 5X freezing problem. It's because of rild. In Halium reference, oFono doesn't start and initialize rild. In Ubuntu Touch, it starts as soon as dbus is up. Feb 10 02:33:59 ril has some freaking crazy control over the system it seems. Feb 10 02:34:20 Ooh Feb 10 02:34:35 So, I replaced the `start` line in `/etc/init/ofono.override` with `manual` and it booted right up. Feb 10 02:35:47 doesn't look like telnet gets started. pretty sure it's panicing Feb 10 02:39:05 I assume telnet isn't somehow integrated into the kernel, right? :p Feb 10 02:39:13 No, that's part of hybris-boot Feb 10 02:43:34 haha I was being overly optimistic Feb 10 02:43:55 https://gist.github.com/rockybulwinkle/4f6f2ad77a3295913840d66bfa407d00 Feb 10 02:44:06 I'm not 100% sure where one log ends and another begins in there. Feb 10 02:44:18 Nothinig a little bit of kernel panic can't fix Feb 10 02:44:54 That is a very unhappy kernel Feb 10 02:45:03 yes. Feb 10 02:47:42 I'm pretty sure I don't have a device tree yet thinking back, breakfast complains "Trying dependencies-only mode on a non-existing device tree?" Feb 10 02:47:50 That's no problem Feb 10 02:51:23 @UniversalSuperBox, That's lie.. Feb 10 02:51:28 You're right Feb 10 02:51:36 ofono doesn't initialize rild Feb 10 02:51:38 I'm back to where I was Feb 10 02:51:40 I get the same error with the sepolicy repo synced, so my device manifest is right? Feb 10 02:51:52 rild is started by container Feb 10 02:52:08 But that said Feb 10 02:52:20 @UniversalSuperBox repo sync system/core Feb 10 02:52:32 And hopefully it will have fixed your issue Feb 10 02:52:39 👀 Feb 10 02:53:25 Building systemimage... t minus 4 minutes. Feb 10 02:55:13 If it doesn't pass me your actual kernel config Feb 10 02:55:31 It works fine on reference, but not ubuntu touch Feb 10 02:55:42 I don't know what's causing it, but there's a clear separation here. Feb 10 02:56:31 (by actual I mean one in $OUT/blabla/.config) Feb 10 02:56:39 oke Feb 10 02:56:52 It's dumping out the sparse image now... Feb 10 02:59:25 @bhushanshah, here it is in the meantime: https://paste.ubuntu.com/=MzBRz22yW7/ **** ENDING LOGGING AT Sat Feb 10 03:00:01 2018