Sleep, laziness, and vibecoding a dangerous mix.

Context

This project is my try on making a old phone a Motorola G9 Plus have the latest Android (16 because 17 is not yet released by LineageOS). In the past had bootloop problems, phone wipes problems, almost bricking it a few thousand times.

In some AI session (likely Claude) a try to fix the bootloop, created a faster bootloop. And since me and Claude, GPT and Kimi been trying to choose a path forward.

Kernel Update

In the last devlog, I said about possible paths forward to fix the BPF (Berkeley Packet Filter), and the easiest solution is to try using OpenEL Assosiation (openela) 4.14.x kernel.

Its going to be easier, plus the backport of the BPF would be painful but it has been done in the past, so I would likely succeed!

Faster Bootloops

I’ve spent all this devlog time, trying to find the culprit and the veridict is… I don’t know, I’ve speding my time looking back at Claude Sessions and I found that it had messed with zygote in one of its tries to fix the bootloop of ~20-30 seconds, and it made it bootloop faster, so…

It also could be other things but the Zygote angles was likely. I’ve researched the whole project, and I couldn’t find anything tracking his changes, so I guess it had messed somewhere deep into the repo, and we couldn’t figure out where.

Next Steps

I’ve decided to delete the whole lineageos directory, and start from “scratch”. Of course I would use the repos in my Github, but at least, I would be in a more stable directory.

I will also do a small rewrite (third time is a charm) to AGENTS.md, with the instruction “tell me all the places where you messed, and use YOUR FUCKING NATIVE TOOLS TO EDIT NOT BASH MTF”.