#LinuxHelp #Gentoo Anyone have any idea what would cause the system to start building kernels that aren't bootable? Following Duxsco's guide, I've been fine since December, but 7.0.2 was the last bootable kernel I've been able to build. The build operation seems to complete successfully, as far as I can tell.

The guide in question: https://www.duxsco.de/

1. Introduction - Gentoo Linux Setup Guide

Gentoo installation using Measured Boot, Secure Boot, Full Disk Encryption, RAID and offering a rescue system based on customised Gentoo cloud image and/or SystemRescueCD

@kewliomzx Which bootloader? What FS are the kernels on?
@ScottE Neither systemd-boot nor rEFInd see the new kernel as bootable, or offer it as an option to even try. The kernels are on an EFI (basically FAT32) partition. 7.0.2 works, but nothing since shows up.

@kewliomzx Hmm. There was a filesystem driver bug in refind drivers that caused the same symptom (newer kernels would get written as sparse files, which broke things), but afaik that's not possible on FAT32.

If you try and launch the kernel directly from the efi shell, what happens?

@ScottE It took me a hot minute to figure out how to do this, but I eventually got "Command error status: Unsupported". I'm pretty sure I did the correct thing, as I was then able to boot the working kernel from the shell.

@kewliomzx So weird.

Wonder if you build a new kernel with the config from the old one if it'll work? (/proc/config.gz or however you can)

Might be some missing EFI stub kernel config?

@ScottE Using the config from 7.0.2 appears to have no discernible effect. The kernel is still not bootable.
@kewliomzx Wtf. Hrmph.
@ScottE yeah I dunno. Starting to think this is something I have to take to either the Gentoo forum or the IRC or something. Dunno how much, or even if, they can help, given I used an alternative setup guide.