| *** ubuntu is now known as zhxt | 01:51 | |
| KayMW | Hi, I'm porting SailfishOS to Nexus 5X (bullhead), and I met a problem at 7.2.1 in HADK (Building the droid-hal-device packages). My phone's SoC is Snapdragon 808 which is AArch64, so I followed the instructions in the output (define droid_target_aarch64) but it failed again at compiling libsparse. Here is the log: http://termbin.com/6i3r . It seems to be another problem about architecture. Any | 04:48 |
|---|---|---|
| KayMW | suggestions? | 04:48 |
| TheKit | KayMW, what base did you use for Nexus 5X? | 06:04 |
| TheKit | you could probably base on ghosalmartin work | 06:05 |
| TheKit | *Android base | 06:05 |
| KayMW | TheKit LineageOS's cm-14.1 branch | 06:05 |
| KayMW | TheKit: His work seems a little out-dated... Anyway, I'll definitely use that as a reference. | 06:10 |
| hacker1024 | Anyone online? | 06:10 |
| TheKit | I'm puzzled what does it need simg2img for actually | 06:10 |
| hacker1024 | My build goes into debug mode on boot | 06:11 |
| hacker1024 | I gen telnet in | 06:11 |
| hacker1024 | can | 06:11 |
| hacker1024 | How do I debug this? | 06:11 |
| hacker1024 | I can''t see anything interesting in init.log or dmesg | 06:11 |
| hacker1024 | The devices name comes up as "refuse..." for a second before changing to a network | 06:12 |
| hacker1024 | \/system is not mounted | 06:14 |
| TheKit | what about /target? | 06:16 |
| hacker1024 | I'll check | 06:17 |
| hacker1024 | Nope | 06:17 |
| hacker1024 | Output of "mount": | 06:18 |
| hacker1024 | /dev/sda15 on / type ext4 (rw,noatime,data=ordered) /dev/sda15 on /data type ext4 (rw,relatime,data=ordered) devpts on /dev/pts type devpts (rw,relatime,mode=600) sysfs on /sys type sysfs (rw,relatime) proc on /proc type proc (rw,relatime) none on /config type configfs (rw,relatime) tmpfs on /dev/shm type tmpfs (rw,nosuid,nodev) tmpfs on /run type tmpfs (rw,nosuid,nodev,mode=755) tmpfs on /sys/fs/cgroup type tmpfs (rw | 06:18 |
| TheKit | please you pastebin | 06:18 |
| hacker1024 | Sorry | 06:18 |
| hacker1024 | Any ides? | 06:18 |
| hacker1024 | ideas | 06:18 |
| TheKit | but it failed to mount SailfishOS rootfs, check init.log for the reason why | 06:18 |
| hacker1024 | I can't see anything | 06:18 |
| TheKit | pastebin it | 06:19 |
| hacker1024 | I'll upload it | 06:19 |
| TheKit | usually the likely reason for this is a mistake with fixup-mountpoints | 06:19 |
| hacker1024 | https://hastebin.com/febehenezo.md | 06:20 |
| hacker1024 | Hmm. I used someone elses mountpoint for this device - I'll double-check them. Thanks. | 06:21 |
| hacker1024 | This could explain why no halium distros are working | 06:21 |
| KayMW | Problem solved by setting USE_STD_C99=0 for libsparse | 06:22 |
| TheKit | hacker1024, are you building SailfishOS or Halium? | 06:22 |
| hacker1024 | I first tried halium, then tried sailfishos | 06:23 |
| hacker1024 | But I used thae same mountpoints for both | 06:23 |
| TheKit | I checked your init.log, disregard about mountpoints | 06:23 |
| TheKit | it did switch to rootfs | 06:23 |
| hacker1024 | Thanks | 06:23 |
| hacker1024 | Do you know what's going wrong? | 06:24 |
| TheKit | systemd should start, does journalctl give you any output? | 06:24 |
| hacker1024 | Nope | 06:24 |
| hacker1024 | I feel like the part where the device cums up as "refusing" for a second when booting is important | 06:25 |
| hacker1024 | *comes | 06:25 |
| TheKit | it can refuse after checking kernel config | 06:26 |
| TheKit | what's the output of ps -Af? | 06:27 |
| hacker1024 | Uploading it | 06:27 |
| hacker1024 | https://hastebin.com/logixiduma.apache | 06:28 |
| TheKit | ok, systemd didn't init properly | 06:28 |
| TheKit | can be an issue with your kernel config, did you use mer_verify_kernel_config on it? | 06:29 |
| hacker1024 | Yes | 06:29 |
| TheKit | cat /proc/config.gz | gzip -dc | 06:29 |
| hacker1024 | cat: /proc/config.gz: No such file or directory | 06:30 |
| hacker1024 | Hmm that should be enabled | 06:30 |
| hacker1024 | Weird | 06:30 |
| TheKit | probably wrong kernel got included into build somehow | 06:31 |
| TheKit | what's your device? | 06:31 |
| KayMW | Opened a new PR for that problem, https://github.com/mer-hybris/droid-hal-device/pull/187 | 06:31 |
| hacker1024 | Xiaomi Mi 5s Plus (natrium) | 06:31 |
| hacker1024 | @KayMW was that directed at me? | 06:31 |
| hacker1024 | @KAYMW I fixed that by using a different implementation of the min macro | 06:32 |
| hacker1024 | @TheKit I'm recompiling the kernel | 06:34 |
| KayMW | hacker1024 wow | 06:35 |
| TheKit | LineageOS 14.1 base? | 06:35 |
| KayMW | Yep | 06:35 |
| hacker1024 | ? | 06:36 |
| TheKit | *that was directed to hacker1024 | 06:36 |
| hacker1024 | I'm confused | 06:36 |
| KayMW | emmmm But I just updated the source several minutes ago, without setting USE_STD_C99 to 0 it still fails... | 06:37 |
| hacker1024 | I didn't push it anywhere... | 06:38 |
| KayMW | emmmmmm 😂 | 06:39 |
| hacker1024 | I can't get the kernel to make the config available through /proc | 06:39 |
| hacker1024 | even though they're enable in the config | 06:39 |
| hacker1024 | CONFIG_IKCONFIG=y CONFIG_IKCONFIG_PROC=y | 06:39 |
| TheKit | are they also enabled in out/target/device/obj/KERNEL_OBJ/.config? | 06:41 |
| hacker1024 | Yep | 06:41 |
| hacker1024 | I'm gonna mka clean and try again | 06:42 |
| TheKit | check how hybris-boot.img is created, either old or prebuilt (but unlikely with your device) kernel is used | 06:42 |
| hacker1024 | I'm building the kernel from LOS sources | 06:42 |
| hacker1024 | But is it possible the kernel's cacheds somewhere? | 06:44 |
| hacker1024 | Got the correct config! but still not booting. | 06:49 |
| hacker1024 | @TheKit https://hastebin.com/xixekosemo.vbs | 06:52 |
| hacker1024 | System mounts now! | 06:52 |
| TheKit | well, that's completely different issue now | 06:52 |
| hacker1024 | Yeah | 06:52 |
| TheKit | systemd starts fine now | 06:52 |
| hacker1024 | journalctl has output | 06:52 |
| hacker1024 | So yes | 06:52 |
| hacker1024 | But I still don't see any errors in any logs... | 06:53 |
| hacker1024 | Running the check on the config from /proc still shows warnings... | 06:55 |
| hacker1024 | It doesn't match up with the config (apparently) used to build | 06:55 |
| hacker1024 | but it says the current value is y and the allowed value is y on every warning... | 06:56 |
| TheKit | well, current issue is likely somewhere else | 07:03 |
| TheKit | systemctl status droid-hal-init | 07:03 |
| hacker1024 | https://hastebin.com/oyivapotot.swift | 07:05 |
| hacker1024 | inactive (dead) | 07:05 |
| TheKit | did you follow hadk faq for hybris-14.1 base? | 07:07 |
| hacker1024 | No - i didn't know there was one | 07:08 |
| hacker1024 | Why? | 07:08 |
| hacker1024 | I didn't know there was a faq I mean | 07:09 |
| TheKit | hacker1024, check the channel's topic | 07:13 |
| hacker1024 | I can't find a FAQ | 07:13 |
| hacker1024 | OK | 07:13 |
| TheKit | https://public.etherpad-mozilla.org/p/faq-hadk | 07:13 |
| hacker1024 | Thanks | 07:13 |
| TheKit | check Notes on LOS14.1 Porting | 07:14 |
| TheKit | , line 657 | 07:14 |
| hacker1024 | Reading | 07:16 |
| hacker1024 | I ran the script | 07:16 |
| hacker1024 | Why does it say, "remove the word exec..." | 07:17 |
| hacker1024 | What do I rebuild now? | 07:18 |
| T4 | <NotKit> droid-hal/droid-config | 07:22 |
| hacker1024 | Thanks | 07:22 |
| hacker1024 | how do I specifically build only droid-config? | 07:32 |
| *** merbot` is now known as merbot | 07:44 | |
| branek_ | Hello... | 07:46 |
| hacker1024 | hello | 07:47 |
| branek_ | For no obvious reason, my HOST starts to trow me an error: File not found: /home/branek/hadk/hybris/mw/libhybris/installroot/usr/lib/libgralloc.so | 07:47 |
| branek_ | same for libgralloc.so.1, and libgralloc.so.1.0.0 | 07:48 |
| hacker1024 | Have you tried syncing again? | 07:48 |
| branek_ | but previous builds have it without problem. | 07:48 |
| hacker1024 | What hybris branch are you using? | 07:48 |
| hacker1024 | No harm in syncing again though... | 07:48 |
| branek_ | https://github.com/mer-hybris/libhybris/tree/master, with changed hybris.spec: https://pastebin.com/xqasMsRj | 07:50 |
| branek_ | I tried with sync... same error... | 07:51 |
| krnlyng | branek_, where did you get that spec from? | 08:40 |
| krnlyng | branek_, libgralloc.so is not part of libhybris, at least not yet | 08:40 |
| hacker1024 | I'm getting this error creating the rootfs: | 08:47 |
| hacker1024 | Error <creator>: URLGrabber error: http://repo.merproject.org/obs/nemo:/devel:/hw:/common/sailfish_latest_aarch64/repodata/repomd.xml - e.errno: 14 | 08:47 |
| hacker1024 | Error 14 is a HTTPError (includes .code and .exception attributes) | 08:48 |
| hacker1024 | nvm port_arch set wrong | 08:55 |
| hacker1024 | My phone is booting into init-debug | 09:23 |
| hacker1024 | Systemd starts, though | 09:23 |
| hacker1024 | droid-hal-init is dead | 09:23 |
| hacker1024 | anyone? | 09:24 |
| ohyash | platform sdk on my side doesn't seem to be complete | 09:36 |
| ohyash | and I'm not sure whats lacking | 09:36 |
| ohyash | there was no tar package. | 09:36 |
| ohyash | I installed it via zypper | 09:36 |
| ohyash | now there's no ubu-chroot | 09:36 |
| r0kk3rz | install android-tools-hadk | 09:38 |
| ohyash | works, thanks | 09:41 |
| branek | mal: any hint/help for missing libgralloc.so missing? I had it on previous build, but I cleaned everything after adding ADRENO_QUIRKS, but now rebuild with build_packages.sh fails with libgralloc missing. | 09:46 |
| hacker1024 | I'm getting this error when trying to run graphics tests: library "libgui.so" wasn't loaded and RTLD_NOLOAD prevented it | 09:49 |
| krnlyng | branek, are you building hybris-15.1? | 09:50 |
| branek | sure? | 09:50 |
| branek | krnlyng: sorry... yes, hybris-15.1 | 09:51 |
| krnlyng | branek, for hybris-15.1 you need android8-initial branch of libhybris not master | 09:51 |
| branek | krnlyng: ah... I used mer-hybris repo instead of yours... thanks... will try now... | 09:56 |
| ohyash | is this normal? https://pastebin.com/6649rf82 | 10:09 |
| ohyash | is asks "Trying dependencies-only mode on a non-existing device tree?" | 10:09 |
| ohyash | but I have all in my local manifests | 10:09 |
| ohyash | uhm, anyone | 10:17 |
| ohyash | ? | 10:18 |
| mal | so it doesmn't build? | 10:18 |
| ohyash | it stops me from trying to build. It looks like it wont build for my device but some generic code because it isnt detecting my device tree. Also, it doesnt list anything like "sailfish" or "hybris" or "2.2.0.xx" in its output. Just got me wondering a little. | 10:22 |
| ohyash | sorry tho, trying to build now | 10:22 |
| mal | android side doesn't know anything about sailfish, you just run breakfast $DEVICE and then make hybris-hal | 10:26 |
| mal | are you again overthinking things? | 10:27 |
| r0kk3rz | or maybe read through the whole hadk to get an idea of the high level process | 10:28 |
| mal | there is some tendency among new porters toward overthinking the whole build process and focusing on all messages instead of just following the instructions | 10:32 |
| r0kk3rz | tbh i prefer that to 'it no work, what do?' new porters | 10:44 |
| mal | true | 10:44 |
| hacker1024 | Getting these errors in logcat, have no graphics. | 10:52 |
| hacker1024 | https://hastebin.com/zosonidaxi.sql | 10:52 |
| *** redj_ is now known as redj | 12:09 | |
| *** lynxis_ is now known as lynxis | 13:39 | |
| jsyreviews | Hi, so I'm having a truoble entering ubu-chrrot, can anyone take a look at this and tell me what did I miss? https://hastebin.com/cemeyupedi.pl | 14:02 |
| jsyreviews | [oh god i can't spell today haha] | 14:02 |
| mal | jsyreviews: install android-tools-hadk | 14:03 |
| mal | in sdk | 14:03 |
| jsyreviews | Okay | 14:03 |
| mal | something like sudo zypper in android-tools-hadk | 14:04 |
| jsyreviews | Yep, installed. | 14:04 |
| jsyreviews | Okay seems like ubu-chrrot works now. | 14:04 |
| jsyreviews | [what is wrong with my spelling today lol] | 14:04 |
| demonaxsh | Hello. I rebuild droidmedia, daemons /usr/libexec/droid-hybris/system/bin/minimediaservice and /usr/libexec/droid-hybris/system/bin/minisfservice worked, but it seems it does not affect situation, fucking screen is black. On strace (jolla-startupwizard-pre-user-session) output all simply: futex(0x740aec, FUTEX_WAIT_PRIVATE, 13, NULL | 14:17 |
| demonaxsh | There is still such a suspicious warning, but perhaps it can not influence anything | 14:20 |
| demonaxsh | [W] unknown:0 - Suspicious physical size QSizeF(4.29497e+09, 4.29497e+09) Check your device configuration! | 14:20 |
| mal | that is not serious issue afaik, but you can define the size in /var/lib/environment/compositor/droid-hal-device.conf like this https://github.com/mlehtima/droid-config-fp2-sibon/blob/master/sparse/var/lib/environment/compositor/droid-hal-device.conf#L6 | 14:22 |
| mal | demonaxsh: is "EGL_PLATFORM=hwcomposer test_hwcomposer" working? | 14:22 |
| mal | demonaxsh: which device was that again? qcom or something else? | 14:23 |
| demonaxsh | not crashed, but sceen black | 14:24 |
| demonaxsh | no, it's mtk device | 14:24 |
| mal | have you check the faq linked in topic and in there section "Devices with Mali GPU" ? | 14:24 |
| demonaxsh | no, but i'll try | 14:26 |
| T4 | <adampigg> Mal, have you worked out how to do 3 different mount commands cleanly? | 15:52 |
| mal | I haven't looked into that yet | 16:03 |
| mal | not completely sure how to handle those except checking kernel version perhaps | 16:03 |
| vknecht | hello | 16:16 |
| jsyreviews | Currently going through setting the $DEVICE.xml file in the /.repo/local_manifests, Goal is porting to S5 klte, using Lineage OS 14.1 as base, can anyone double check this? https://hastebin.com/utigokases.xml | 16:25 |
| mal | jsyreviews: is there a good reason you use the old cyanogenmod repos and not the newer lineageos repos? | 16:31 |
| jsyreviews | Nope, This is just a plain edit, hence the reason why I dropped it here to let you guys double check first. | 16:32 |
| mal | jsyreviews: also you are missing some repos, you forgot to check the cm or lineage.dependencies files | 16:34 |
| mal | so for example you have https://github.com/LineageOS/android_device_samsung_klte/blob/cm-14.1/lineage.dependencies and then you check that repo for similar dependency file and so on until there are no dependency files anymore | 16:35 |
| jsyreviews | Okay. | 16:35 |
| mal | that repo listed there has https://github.com/LineageOS/android_device_samsung_klte-common/blob/cm-14.1/lineage.dependencies and so on | 16:35 |
| mal | you need to add all of the repos listed in those files | 16:36 |
| mal | and in the files you find in the dependency repos | 16:36 |
| mal | so quite many repos | 16:36 |
| jsyreviews | Noted, once I looked into all of them, i'll drop another haste bin just to make sure | 16:36 |
| piggz | mal: yeah, did my version read /proc/filesystems? you could do something similar | 20:20 |
| Umeaboy | Hi! I found a working setup to build a unofficial lineage-15.1 zip for my phone, but it fails at 80% because I lack of more memory even if I increase the java heap size to 6GB. Is there some here with more than 8GB who's willing to build for me? | 23:08 |
| Umeaboy | This could be my chance to try it out and then build hybris. | 23:09 |
Generated by irclog2html.py 2.17.1 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!