**** BEGIN LOGGING AT Fri Jan 04 02:59:57 2019
Jan 04 09:13:03 <ubptgbot> <Nebrassy> so I got the container working but I get this error in systemctl status lxc@android … c-start[1327]: lxc-start: start.c: preserve_ns: 153 No such file or directory - Kernel does not support attaching to pid namespaces.
Jan 04 09:13:10 <ubptgbot> <Nebrassy> should I worry about that?
Jan 04 09:17:27 <ubptgbot> <Mattia990> @Nebrassy, Check the lxc-checkconfig
Jan 04 09:17:40 <ubptgbot> <Mattia990> @Nebrassy, [Edit] Check the lxc-checkconfig command
Jan 04 09:27:48 <ubptgbot> <Nebrassy> @Mattia990, https://paste.ubuntu.com/p/wnM955Bx55/
Jan 04 09:28:33 <ubptgbot> <Nebrassy> I have enabled everything I could, ones that are still missing don't have an entry in m7_defconfig, but I notice user namespace is enabled
Jan 04 09:37:20 <ubptgbot> <Mattia990> @Nebrassy, I think the problem may be user namespaces
Jan 04 09:38:54 <ubptgbot> <Nebrassy> @Mattia990, What's the config to disable that
Jan 04 09:40:14 <ubptgbot> <Mattia990> @Nebrassy, I think this: CONFIG_USER_NS
Jan 04 09:49:00 <ubptgbot> <rogieroudshoorn> isn't user namespaces supposed to be enabled as per LXC guidelines?
Jan 04 09:50:54 <ubptgbot> <Nebrassy> @rogieroudshoorn, According to https://docs.halium.org/en/latest/porting/debug-build/debug-android-userspace.html … All option except User namespace need to be the green word enabled. If one of the options is a yellow missing or a red required, then you need to change the kernel config, rebuild hybris-boot and check the statu
Jan 04 09:50:54 <ubptgbot> s again.
Jan 04 09:51:20 <ubptgbot> <rogieroudshoorn> ah right, thnx
Jan 04 09:51:23 <ubptgbot> <rogieroudshoorn> i'
Jan 04 09:51:29 <ubptgbot> <rogieroudshoorn> should better double check that :)
Jan 04 09:53:45 <ubptgbot> <Nebrassy> failing to build halium-boot … https://paste.ubuntu.com/p/p7Tqs8Y6BQ/
Jan 04 10:16:35 <ubptgbot> <rogieroudshoorn> crap, i got user namespace enabled on my s4 mini - that's probably what was breaking libhybris on pmos
Jan 04 10:23:43 <ubptgbot> <Nebrassy> I still get … lhost.localdomain lxc-start[1310]: lxc-start: start.c: preserve_ns: 153 No such file or directory - Kernel does not support attaching to pid namespaces. … after disabling user namespace, and halium-boot still fails to build
Jan 04 10:27:05 <ubptgbot> <rogieroudshoorn> NS_PID is it's own configuration parameter, do you have it set?
Jan 04 10:28:46 <ubptgbot> <Nebrassy> CONFIG_PID_NS=y … yes
Jan 04 10:29:12 <ubptgbot> <rogieroudshoorn> are you using make menuconfig for your kernel config editing?
Jan 04 10:29:49 <ubptgbot> <Nebrassy> No just a text editor
Jan 04 10:30:07 <ubptgbot> <rogieroudshoorn> kernel configuration flags have internal dependencies; make menuconfig handles those for you
Jan 04 10:30:24 <ubptgbot> <rogieroudshoorn> you may not be building the kernel with the flags you have this way
Jan 04 10:36:15 <ubptgbot> <rogieroudshoorn> i have another question - should i be using halium-boot or hybris-boot?
Jan 04 10:37:11 <ubptgbot> <Nebrassy> @rogieroudshoorn, depends on what you want it for
Jan 04 10:37:30 <ubptgbot> <rogieroudshoorn> right now? just plain halium
Jan 04 10:37:39 <ubptgbot> <Nebrassy> I'm not sure how to use make menuconfig, what's the proper way to do that
Jan 04 10:37:59 <ubptgbot> <Nebrassy> @rogieroudshoorn, use hybris-boot initially, you'll need halium-boot for ubuntu touch later
Jan 04 10:38:24 <ubptgbot> <Sergobot> @Nebrassy, Go to your kernel directory and run it there
Jan 04 10:39:40 <ubptgbot> <Nebrassy> @Sergobot, tried doing that but I keep getting errors, do I need to run breakfast and such first?
Jan 04 10:40:00 <ubptgbot> <Sergobot> Hmm, I don't think so
Jan 04 10:40:03 <ubptgbot> MarcoDS_bit was added by: MarcoDS_bit
Jan 04 10:40:05 <ubptgbot> <Nebrassy> https://paste.ubuntu.com/p/2t7XFXZx3k/
Jan 04 10:40:09 <ubptgbot> <Sergobot> It worked just fine for me
Jan 04 10:41:00 <ubptgbot> <rogieroudshoorn> if you're based on lineage, there's probably a shell script to edit it in that directory (there was for my s4 mini)
Jan 04 10:43:20 <ubptgbot> <minlexx> @Sergobot, ARCH=arm CROSS_COMPILE=arm-whatever-that-triplet-is- make menuconfig
Jan 04 10:43:52 <ubptgbot> <Nebrassy> @rogieroudshoorn, got the same errors running that script
Jan 04 10:44:21 <ubptgbot> <Sergobot> @minlexx, Won't it compile for arm-something? The script is executed on host
Jan 04 10:47:27 <ubptgbot> <minlexx> @Nebrassy, Wrong message to reply, meant this @Sergobot 😊 … That's the command to invoke kernel build used in Halium (without menuconfig)
Jan 04 10:48:43 <ubptgbot> <Nebrassy> @Sergobot, the script does that, anyway it gave an error but I believe it restored the original defconfig
Jan 04 10:51:21 <ubptgbot> <Nebrassy> anyway, anyone got an idea why halium-boot is failing?
Jan 04 10:57:08 <ubptgbot> <Dzmitry Sankouski> @minlexx, ARCH=arm CROSS_COMPILE=arm-whatever-that-triplet-is- KCONFIG_CONFIG=path-to-config-file-you-want-edit make menuconfig
Jan 04 10:59:12 <ubptgbot> <rogieroudshoorn> @lyudmila_rev1917 i got a moto g4 athena to play around with, and i started compiling & building your port. i got a few questions though, it seems as if it's halfway to an f2fs implementation for instance ...
Jan 04 11:01:04 <ubptgbot> <minlexx> @Nebrassy, `make O=/home/nebrassy/halium/out/target/product/m7/obj/KERNEL_OBJ ARCH=arm CROSS_COMPILE=arm-linux-androidkernel- menuconfig`
Jan 04 11:01:54 <ubptgbot> <minlexx> @Nebrassy, Parameters are the same as from this paste
Jan 04 11:04:42 <ubptgbot> <minlexx> The biuld fail probably because there is a warning during compilation and warnings are treated as errors
Jan 04 11:05:16 <ubptgbot> <minlexx> So use your C knowledge and fix warning
Jan 04 11:06:08 <ubptgbot> <Nebrassy> @minlexx, https://paste.ubuntu.com/p/DKsqgcDxS4/
Jan 04 11:06:15 <ubptgbot> <Nebrassy> @minlexx, haha what's that
Jan 04 11:06:53 <ubptgbot> <minlexx> ncurses-dev is needed
Jan 04 11:09:25 <ubptgbot> <minlexx> @Nebrassy, The required skill to do porting? 😁
Jan 04 11:12:11 <ubptgbot> <NotKit> @rogieroudshoorn, unlikely
Jan 04 11:12:49 <ubptgbot> <NotKit> what is the device and what you tried on pmos?
Jan 04 11:13:09 <ubptgbot> <rogieroudshoorn> device is s4 mini duos
Jan 04 11:13:34 <ubptgbot> <rogieroudshoorn> i built full halium, booted into plasma rootfs (with visuals), downloaded the systemimage, and pushed that into my pmos port of the device
Jan 04 11:13:49 <ubptgbot> <rogieroudshoorn> errors i get smell like caf headers though
Jan 04 11:13:56 <ubptgbot> <NotKit> Snapdragon 400, ok
Jan 04 11:13:56 <ubptgbot> <rogieroudshoorn> but i thought you fixed those
Jan 04 11:14:16 <ubptgbot> <NotKit> is Halium 7.1 or 5.1 used as base?
Jan 04 11:14:18 <ubptgbot> <rogieroudshoorn> so presumed it must be my port
Jan 04 11:14:22 <ubptgbot> <rogieroudshoorn> it's based on 7.1
Jan 04 11:14:33 <ubptgbot> <rogieroudshoorn> lineageos versions are available up to pie even
Jan 04 11:14:39 <ubptgbot> <NotKit> test_hwcomposer_7.1_caf did not work?
Jan 04 11:15:50 <ubptgbot> <rogieroudshoorn> no, is there a different command for it then test_hwcomposer?
Jan 04 11:16:07 <ubptgbot> <rogieroudshoorn> to me it seemed it was grabbing a libhybris version it didn't need though at apk install\
Jan 04 11:16:21 <ubptgbot> <NotKit> after recent packaging update another binary is provided which is built with CAF changes
Jan 04 11:16:32 <ubptgbot> <rogieroudshoorn> it also installed libhybris-7.1 regular as a dependency
Jan 04 11:16:36 <ubptgbot> <rogieroudshoorn> ah right
Jan 04 11:17:11 <ubptgbot> <rogieroudshoorn> so reinstall of libhybris packages from the device will fix that?>
Jan 04 11:18:20 <ubptgbot> <NotKit> try running EGL_PLATFORM=hwcomposer test_hwcomposer_7.1_caf
Jan 04 11:19:29 <ubptgbot> <NotKit> [Edit] try running EGL_PLATFORM=hwcomposer test_hwcomposer_7_caf
Jan 04 11:19:44 <ubptgbot> <NotKit> [Edit] test_hwcomposer_7_caf did not work?
Jan 04 11:20:09 <ubptgbot> <rogieroudshoorn> yeah, that works
Jan 04 11:22:52 <ubptgbot> <rogieroudshoorn> awesome, thank you TheKit! (and let's continue this on the postmarketOS channel i guess )
Jan 04 11:26:17 <ubptgbot> <rogieroudshoorn> one more question for this group, what will break or go wrong due to the user namespace activated?
Jan 04 11:36:03 <ubptgbot> <NotKit> it is not an issue to have it activated, the docs don't suggest it because many vendor kernels require fixes to build with it activated
Jan 04 11:36:47 <ubptgbot> <NotKit> user namespace is used to have non-priveleged (run as user) LXC containers, which we don't utilize
Jan 04 11:38:56 <ubptgbot> <NotKit> https://wiki.archlinux.org/index.php/Linux_Containers#Privileged_containers_or_unprivileged_containers - if interested, read this
Jan 04 11:39:25 <ubptgbot> <NotKit> [Edit] https://wiki.archlinux.org/index.php/Linux_Containers#Privileged_containers_or_unprivileged_containers - if interested, read this, Halium uses privileged container
Jan 04 11:40:02 <ubptgbot> <rogieroudshoorn> ah, right.
Jan 04 11:40:06 <ubptgbot> <rogieroudshoorn> thank you :)
Jan 04 11:51:03 <ubptgbot> <Nebrassy> @NotKit, but it seems like halium-boot does, the script halium-boot/check-kernel-config activates it
Jan 04 11:51:47 <ubptgbot> <Nebrassy> I think this is the error causing the fail, but I don't know what's the problem, source file looks fine … /home/nebrassy/halium/kernel/htc/msm8960/fs/ecryptfs/keystore.c: In function 'decrypt_pki_encrypted_session_key': … /home/nebrassy/halium/kernel/htc/msm8960/fs/ecryptfs/keystore.c:1168:7: warning: 'payload_len' may be used unin
Jan 04 11:51:47 <ubptgbot> itialized in this function [-Wmaybe-uninitialized] … error, forbidden warning: keystore.c:1168
Jan 04 12:06:42 <ubptgbot> <Dzmitry Sankouski> You can disable warnings. See http://gcc.gnu.org/onlinedocs/gcc/Warning-Options.html … You may  add -Wno-maybe-uninitialized  options to compiler's options in makefile responsible for building this file
Jan 04 12:45:46 <ubptgbot> <rogieroudshoorn> my dmesg is spammed by errors such as : Service 'vold' (pid 11333) killed by signal 6
Jan 04 12:45:55 <ubptgbot> <rogieroudshoorn> i can't find anything in logcat
Jan 04 13:23:44 <ubptgbot> <lyudmila_rev1917> @rogieroudshoorn, the f2fs patch was merged
Jan 04 13:51:02 <ubptgbot> <rogieroudshoorn> @lyudmila_rev1917, well, i get stuck in initrd with this f2fs error: [    3.484482,4] initrd: checking filesystem integrity for the userdata partition … [    3.489532,4] F2FS-fs (mmcblk0p48): Unrecognized mount option "errors=remount-ro" or missing value … [    3.489578,4] F2FS-fs (mmcblk0p48): Unrecognized mount optio
Jan 04 13:51:03 <ubptgbot> n "errors=remount-ro" or missing value … [    3.489850,4] mount: mounting /dev/mmcblk0p48 on /tmpmnt failed: Invalid argument
Jan 04 13:51:41 <ubptgbot> <rogieroudshoorn> and i'm trying to figure out where this is done ... maybe you can point me to the right direction here?
Jan 04 13:51:51 <ubptgbot> <lyudmila_rev1917> @rogieroudshoorn, should be mounted
Jan 04 13:52:13 <ubptgbot> <Mattia990> @rogieroudshoorn, Remove that argument from fstab.qcom (maybe)
Jan 04 13:52:20 <ubptgbot> <Mattia990> @rogieroudshoorn, [Edit] Remove that arguments from fstab.qcom (maybe)
Jan 04 13:54:30 <ubptgbot> <rogieroudshoorn> yeah, it's not in there :)
Jan 04 13:54:30 <ubptgbot> <lyudmila_rev1917> @Mattia990, its not even there
Jan 04 13:54:58 <ubptgbot> <rogieroudshoorn> perhaps i should reformat /data on f2fs in twrp?
Jan 04 13:55:02 <ubptgbot> <rogieroudshoorn> that'll probably help
Jan 04 13:55:59 <ubptgbot> <rogieroudshoorn> what is there, is two declarations for data; one f2fs and one ext4 - is that how it's supposed to be done?
Jan 04 13:56:48 <ubptgbot> <lyudmila_rev1917> @rogieroudshoorn, theres two entries because theres two entries
Jan 04 13:56:58 <ubptgbot> <lyudmila_rev1917> @lyudmila_rev1917, check if its even mounted
Jan 04 13:57:20 <ubptgbot> <rogieroudshoorn> no, it doesn't mount it
Jan 04 13:57:28 <ubptgbot> <rogieroudshoorn> target dir is empty
Jan 04 13:57:45 <ubptgbot> <lyudmila_rev1917> @rogieroudshoorn, ls /data
Jan 04 14:00:03 <ubptgbot> <rogieroudshoorn> battery drained now; i'll need to charge it atm
Jan 04 14:08:09 <ubptgbot> <rogieroudshoorn> managed to boot it; no data
Jan 04 14:08:17 <ubptgbot> <rogieroudshoorn> (it's in initrd)
Jan 04 14:09:15 <ubptgbot> <lyudmila_rev1917> @rogieroudshoorn, and why is it on initrd?
Jan 04 14:09:32 <ubptgbot> <rogieroudshoorn> because it gets stuck there
Jan 04 14:09:50 <ubptgbot> <lyudmila_rev1917> @rogieroudshoorn, and why is it stuck there?
Jan 04 14:10:05 <ubptgbot> <rogieroudshoorn> best guess, because of the mount errors
Jan 04 14:11:22 <ubptgbot> <rogieroudshoorn> full dmesg: https://paste.ubuntu.com/p/4vqH7Rw8fd/
Jan 04 14:14:24 <ubptgbot> <lyudmila_rev1917> @rogieroudshoorn, [    3.485900,4] F2FS-fs (mmcblk0p48): Unrecognized mount option "data=journal" or missing value … [    3.485936,4] F2FS-fs (mmcblk0p48): Unrecognized mount option "data=journal" or missing value … [    3.486162,4] mount: mounting /dev/mmcblk0p48 on /tmpmnt failed: Invalid argument
Jan 04 14:16:07 <ubptgbot> <rogieroudshoorn> yeah; i'm trying to figure out where those mount options & arguments are pushed in
Jan 04 14:16:19 <ubptgbot> <NotKit> wasn't the fix supposed to prevent that?
Jan 04 14:16:26 <ubptgbot> <lyudmila_rev1917> @NotKit, yep
Jan 04 14:16:40 <ubptgbot> <lyudmila_rev1917> but he didnt sync or it didnt download the needed initrd
Jan 04 14:20:18 <ubptgbot> <rogieroudshoorn> i synced it on tuesday
Jan 04 14:22:13 <ubptgbot> <NotKit> find out -name halium-initramfs.gz
Jan 04 14:22:31 <ubptgbot> <NotKit> if it finds, -delete and try to rebuild
Jan 04 14:24:22 <ubptgbot> <rogieroudshoorn> found it, trying
Jan 04 14:25:09 <ubptgbot> <rogieroudshoorn> [ 50% 5/10] Downloading initramfs to : /home/rogier/Development/halium/out/target/product/athene/obj/ROOT/halium-boot_intermediates/halium-initramfs.gz … Downloading https://github.com/halium/initramfs-tools-halium/releases/download/continuous/initrd.img-touch-armhf
Jan 04 14:25:14 <ubptgbot> <rogieroudshoorn> is that what you expected?
Jan 04 14:26:13 <ubptgbot> <NotKit> yes
Jan 04 14:26:30 <ubptgbot> <lyudmila_rev1917> it downloads the older one
Jan 04 14:27:06 <ubptgbot> <NotKit> shouldn't this fetch latest release?
Jan 04 14:27:11 <ubptgbot> <NotKit> https://github.com/halium/initramfs-tools-halium/releases/
Jan 04 14:27:24 <ubptgbot> <lyudmila_rev1917> @NotKit, should
Jan 04 14:27:49 <ubptgbot> <NotKit> hm, yes, you're right, continuous is not updated
Jan 04 14:28:22 <ubptgbot> <NotKit> @JBBgameich do you know how that tag should be updated?
Jan 04 14:28:35 <ubptgbot> <rogieroudshoorn> yeah, feb 2018
Jan 04 14:28:40 <ubptgbot> <rogieroudshoorn> slow n00b here, sorry ;)
Jan 04 14:28:53 <ubptgbot> <lyudmila_rev1917> 7 feb...
Jan 04 14:29:04 <ubptgbot> <lyudmila_rev1917> we have been using a ramfs of 7feb of 18
Jan 04 14:29:09 <ubptgbot> <JBBgameich> @NotKit, It was supposed to delete the continous tag and add a new one for the latest commit, so the download link keeps working
Jan 04 14:29:52 <ubptgbot> <JBBgameich> But it seems to have stopped working recently. I'm not sure if it's due to a change in our setup or by travis-ci.
Jan 04 14:30:46 <ubptgbot> <NotKit> someone with force-push rights could bump it for now
Jan 04 15:00:43 <ubptgbot> yurikoles was added by: yurikoles
Jan 04 15:08:56 <ubptgbot> amerpeacemaster was added by: amerpeacemaster
Jan 04 15:15:32 <ubptgbot> vpzom was added by: vpzom
Jan 04 23:03:54 <ubptgbot> <Nebrassy> @Dzmitry Sankouski, Thanks this worked, had to add it in the main make file for the kernel under kbuild_flags … Then had another error but found a random obscure 6 years old commit that fixes it
Jan 04 23:04:59 <ubptgbot> <Nebrassy> I should probably fork the repositories at this point, made many changes
Jan 05 02:08:58 <ubptgbot> <Nebrassy> alright, halium-boot fails to boot, can only access via telnet, are there any steps to diagnose?
Jan 05 02:13:53 <ubptgbot> <UniversalSuperBox> Type `dmesg` and find the reason for the panic
Jan 05 02:13:58 <ubptgbot> <UniversalSuperBox> It'll be prefaced by PANIC
Jan 05 02:14:06 <ubptgbot> <lyudmila_rev1917> there should be also a diagnosis.log
Jan 05 02:14:12 <ubptgbot> <UniversalSuperBox> No, it's halium-boot
Jan 05 02:14:17 <ubptgbot> <lyudmila_rev1917> oh right
Jan 05 02:16:19 <ubptgbot> <Nebrassy> [    2.852539] initrd: PANIC for reason: Couldn't find data partition.
Jan 05 02:17:57 <ubptgbot> <Nebrassy> might have an idea why's that, in fstab I have two entries for data, one is f2fs and the other is ext4, but my data partition is ext4, could that be the reason?
Jan 05 02:27:00 <ubptgbot> <Tygerpro Tygerpro> On my port I had to manually set the data partition (as well as the rest of them) in my fstab file (I.e. /dev/mmcblk0p12345)
Jan 05 02:28:41 <ubptgbot> <Nebrassy> I've edited the fixup-mountpoints script in halium/hybris-boot, does halium-boot not use that?
Jan 05 02:29:06 <ubptgbot> <Nebrassy> also using telnet I've cat /etc/fstab and it's empty
Jan 05 02:29:40 <ubptgbot> <UniversalSuperBox> It is supposed to find the partition automatically if it is labeled correctly, but that could not happen
Jan 05 02:29:44 <ubptgbot> <UniversalSuperBox> What is the partition's name in your fstab?
Jan 05 02:29:50 <ubptgbot> <Nebrassy> userdata
Jan 05 02:30:01 <ubptgbot> <UniversalSuperBox> It should be found, but try Tygerpro's idea first
Jan 05 02:30:16 <ubptgbot> <Nebrassy> alright
Jan 05 02:31:15 <ubptgbot> <Tygerpro Tygerpro> I don't like my solution but it worked in my case
**** ENDING LOGGING AT Sat Jan 05 02:59:56 2019