#RabbitHole: @okias@floss.social is doing a lot work to mainstream #sdm845 patches to the #linux kernel. During a meeting of Phosh.mobi e.V. members I learned that there seem to be not many testers for #Shift6mq #axolotl. I still had a broken Axolotl sitting on my desk…

• repaired the phone that originally had been bought as part of a #ShiftPhone8 deal
  • cleaning inside (residue from water contact)
  • new display (two lines of broken pixels)
  • new usb-c pcb (didn't charge, data broke during testing)
• learned about edl to recover if phone wouldn't boot to fastboot anymore (yes, really had that situation later )
• learned about the difference of fastboot and fastbootd mode
• learned about u-boot and the boot chain on the phone
• learned about dynamic partitions used in Android and how to use them from #linux
• learned about pmbootstrap
  • to build my own device images (mixture of existing axolotl and sdm845 to be compatible to the phone and the usage with u-boot and dynamic partitions)
  • changed pmbootstrap to build small (92M) boot image for boot_a/b partition
• tried to get more space for out of the phone for boot & root than provided by boot_a/b and system_a/b
  • resized userdata of Android
  • created two additional partitions for postmarketOS
• learned how to recover from kernel not booting without re-flashing
  • u-boot usb mass storage mode exports the mass storage of the phone over usb-c
  • phone connected to an aarch64 system (#Librem5 running #Mobian)
  • mounting encrypted partition, boot and what-else is needed
  • installing kernel packages via apk while in chroot on Librem5
• learned how to compile a kernel using ./envkernel.sh and pmbootstrap

Now I'm quite sure I can recover from any failure I will encounter while testing sdm845 kernels and I have the following setup on my Shift6mq:

• ShiftOS-L with small userdata partition in slot_b
  • applications for demoing the phone installed
  • work profile to contain closed source applications
• postmarketOS edge installed on partitions behind userdata
  • u-boot on rescue_a
  • 1GB pmOS_boot
  • 60GB pmOS_root
  • applications for demoing
  • multi-user: my testing user, demo user
  • everything except cameras work
• dual-boot possible on-the-go: enter fastboot (by holding vol-up at start), change slot, reboot

I put my notes on the postmarketOS wiki.

I started two merge requests:

• support audio configuration when started through u-boot in alsa
• add support for booting using u-boot and dynamic partitions to axolotl

I'll carry the phone at #clt2026 - Chemnitzer Linux Tage 2026 - on saturday 28th this month. I'll mostly be around the "Linux on Mobile" booth.

Follow-up to this note:

At one of the times the current #Armbian #linux kernel didn't crash, but booted on my #RockPro64 I installed #GotoSocial and yes: no more strange error messages running it. The only time I remember that I updated a linux kernel to make an application work.

Preparations to get the production system running with the newer kernel:

• in case of problems I'd need to know my way back to boot the old kernel: u-boot recap and learning
• test suspected problems with the #sata #pcie ctrl beforehand (get a second Pine64 sata ctrl for the test system)

Meanwhile Dragan had patched the device tree and asked to check on a few reboots whether this would make a difference or cause any regressions.

Further down the #RabbitHole the tunnel forks…

On my #yunohost I tried to update #gotosocial from 0.17.4~ynh1 to 0.18.1~ynh1 and the update failed.

GotoSocial just showed some cryptical error messages when started.

@dumpsterqueer@superseriousbusiness.org traced the problem back to a changed lib version used in the new GotoSocial version and the developer of the lib answered my #linux kernel would be too old.

The old #armbian legacy kernel is running, because … I don't remember.

I need a kernel update for my #rockpro64.

Down the #RabbitHole…