Sunday, 2021-03-28

T42<mintphin> `tar: exec bunzip2: Too many symbolic links encountered00:13
T42<mintphin> tar: read error``00:13
T42<mintphin> ok i fixed it, i had to use https://github.com/sailfish-lavender/hybris-installer01:27
T42<mintphin> but now there's a new issue01:43
T42<mintphin> when booting it'll immediately go to fastboot01:43
T42<mintphin> forgot to disable initramfs skipping01:54
T42<mintphin> so time to recompile my kernel ig01:54
T42<Ivan %lastname%> @hacker1245502:46
ThaodanWill restart the server now02:55
T42<hacker12455> @Ivan %lastname% [@hacker12455], Yes?06:29
*** techgeekster_ is now known as techgeekster10:23
T42<simonschmeisser> Is presage text prediction within the scope of this channel? I installed from openrepos and it shows up in text input config in settings where I selected "English" and "German (presage)" but the maalit keyboard only shows "EN"13:28
T42Stellar_Eagle was added by: Stellar_Eagle15:22
T42<Stellar_Eagle> @mintphin [question is, how could i get bunzip2 here?], Did you mount system partition in TWRP?15:23
T42<Stellar_Eagle> Updater script doesn't work via TWRP.15:25
T42<Stellar_Eagle> adb push <dir>whatever-rootfs.tar.bz2 /data15:40
T42<Stellar_Eagle> <dir> wherever it is.15:40
T42<Stellar_Eagle> rm -rf /data/.stowaways/sailfishos15:40
T42<Stellar_Eagle> mkdir -p /data/.stowaways/sailfishos15:40
T42<Stellar_Eagle> tar --numeric-owner -xvjf /data/whatever-rootfs.tar.bz2 -C /data/.stowaways/sailfishos15:40
T42<Stellar_Eagle> rm /data/whatever.tar.bz215:40
T42<Stellar_Eagle> Alternately15:40
T42<Stellar_Eagle> tar --numeric-owner -xvjf /data/whatever-rootfs.tar.bz2 \15:40
T42<Stellar_Eagle> > -C /data/.stowaways/sailfishos15:40
T42<Stellar_Eagle> when not on full screen.15:40
T42<mintphin> don't worry, i fixed the issue15:53
T42<mintphin> now my concerns are with the fact it boots directly to fastboot15:53
T42<mintphin> yeah, i tried disabling initramfs skipping15:55
T42<mintphin> that didn't work at all15:55
T42<elros34> any sign of life in host dmesg?15:58
T42<mintphin> lemme check15:58
T42<mintphin> ```[113135.211409] usb 1-5: new high-speed USB device number 100 using ehci-pci16:00
T42<mintphin> [113135.341618] usb 1-5: New USB device found, idVendor=18d1, idProduct=d00d, bcdDevice= 1.0016:00
T42<mintphin> [113135.341629] usb 1-5: New USB device strings: Mfr=1, Product=2, SerialNumber=316:00
T42<mintphin> [113135.341634] usb 1-5: Product: Android16:00
T42<mintphin> [113135.341638] usb 1-5: Manufacturer: Google16:00
T42<mintphin> [113135.341641] usb 1-5: SerialNumber: 142796c43d31```16:00
T42<mintphin> https://github.com/Sailfish-OS-Mi-A3/android_kernel_xiaomi_laurel_sprout/commit/a26eb0af6b702d87e35e155038e39d26aa2b240016:01
T42<mintphin> https://github.com/Sailfish-OS-Mi-A3/android_kernel_xiaomi_laurel_sprout/commit/e4698b4be31064c9c9c5961dfc77321f5f5d1a2f16:01
T42<mintphin> (btw i have disabled QTAGUID after this commit)16:02
T42<elros34> did you flash hybris-boot.img after making skip_initramfs changes?16:03
T42<mintphin> yeah16:04
T42<mintphin> could be the fact i haven't set the SELinux boot parameter to 016:05
T42<elros34> I ask because your dmesg looks like android boot not hybris-boot initramfs. Is that log really from first boot after flasing hybris-boot.img or after device rebooted several times16:06
T42<elros34> no, selinux doesn't matter if initrafms doesn't work16:06
T42<mintphin> after the device has rebooted for several times16:06
T42<mintphin> how can i do a deep shutdown from fastboot though?16:06
T42<elros34> not sure, it might be device specific. IIRC for me it was enough to press power button16:07
T42<mintphin> i did that16:07
T42<mintphin> (held the power button till the device rebooted)16:08
T42<elros34> press not hold. Anyway flash hybris-boot.img, let it reboot and watch dmesg until it reboots16:08
T42<mintphin> oh another method i've used to exit fastboot is leave the device unplugged16:09
T42<mintphin> it completely shuts down after it16:09
T42<elros34> sorry I need to go now16:10
T42<mintphin> ok that's fine!16:10
deathmist_@mintphin does "dmesg | grep Mer" show anything usb related? if you've had the device plugged in while trying to boot using hybris-boot you've not yet reached the initramfs16:16
T42<mintphin> @deathmist_ [@mintphin does "dmesg | grep Mer" show anythin …], nope16:17
deathmist_ok, and this was 17.1 correct? in that case have you "export TEMPORARY_DISABLE_PATH_RESTRICTIONS=true" before building hybris-boot? without that the initramfs cannot be added to the boot image16:18
T42<mintphin> ohhhh yeah it's 17.116:18
T42<mintphin> i didn't know that16:18
T42<mintphin> ok i'm gonna try doing that and i'll get back at you in about an hour16:18
deathmist_you don't have to rebuild everything if that's what you're thinking, just run the export and "mka hybris-boot"16:19
T42<mintphin> ah ok16:19
deathmist_(and you can also just flash/boot the new hybris-boot.img from the out/target/product/$DEVICE/)16:19
T42<mintphin> yeah that's what i did before16:20
T42<mintphin> the file grew significantly in size after doing that but it's still booting directly to fastboot16:48
deathmist_@mintphin yeah it should've grown since now it actually has the ramdisk included, anything new in "dmesg -w" / mention of "Mer"?16:48
T42<mintphin> nope16:49
deathmist_I don't know if this applies to 4.14 kernels, but you can try reverting a few commits as I've done for my 4.4 kernel: https://irclogs.sailfishos.org/logs/%23sailfishos-porters/%23sailfishos-porters.2021-02-17.log.html#t2021-02-17T13:18:5216:49
T42<hacker12455> kacper.grobelny6*XwoRgHr6c8Rg17:14
T42<mintphin> those patches sadly didn't do anything to fix my issue17:36
T42<elros34> @mintphin did you ever need these patches? I briefly checked your kernel previously and did not notice breaking "workqueue to be kthread" commit18:02
T42<elros34> Maybe you can try to reboot directly to twrp and grab /proc/last_kmsg or /sys/fs/pstore18:06
T42<mintphin> aight18:13
T42<Ivan %lastname%> @hacker12455 [Yes?], Do you still have the Galaxy tab 3?18:52
T42<hacker12455> @Ivan %lastname% [Do you still have the Galaxy tab 3?], yes18:52
T42<hacker12455> why?18:52
T42<mintphin> @elros34 [Maybe you can try to reboot directly to twrp a …], they're empty18:59
T42<Ivan %lastname%> @hacker12455 [why?], did you manage to have Sailfish on the device?19:00
T42Brianrh was added by: Brianrh19:03
T42<elros34> @mintphin ok. I can only advice you to revert all commits you have made except obviously skip_initramfs. In defconfig enable only these required by hybris-boot init script except CONFIG_VT. Don't forget to export variable before building kernel19:04
T42<mintphin> still going directly to fastboot20:07
T42<elros34> so what kernel changes do you have now?20:10
T42<mintphin> just skip_initramfs for now20:10
T42<elros34> what about defconfig changes?20:10
T42<mintphin> just disabled CONFIG_VT, the rest is as-is on github20:11
T42<elros34> you were supposed to use default android defconfig without all your changes then add only these (except VT): https://github.com/mer-hybris/hybris-boot/blob/master/init-script#L325-L33020:12
T42<mintphin> oh ok20:14
T42<hacker12455> @Ivan %lastname% [did you manage to have Sailfish on the device?], did not try20:24
T42<hacker12455> you could attempt with cm1120:24
T42<Ivan %lastname%> I want to try to upload the android version but there are kernel errors20:26
T42<mintphin> @elros34 [you were supposed to use default android defco …], same issue21:14
T42<mintphin> (Document) https://irc.thaodan.de/.imgstore/IqTBLEP6sG.txt21:14
T42<zinstack625> @mintphin please use pastebin, don't make people download files21:16
T42<mintphin> ohok21:17
T42<elros34> so basically android kernel still fail to fastboot? Are you sure you exported needed variable before building21:17
T42<mintphin> @elros34 [so basically android kernel still fail to fast …], it boot straight to fastboot21:17
T42<elros34> you can of course  try without any defconfig changes but unlikely any  these options cause issue so something else must be wrong21:24
T42<mintphin> ok21:53
T42<mintphin> uh i'm gonna try not changing anything in defconfig first21:53
T42<mintphin> it still boots straight to fastboot23:00

Generated by irclog2html.py 2.17.1 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!