| Summary: | sys-kernel/gentoo-sources-6.12.16 hangs on boot with loglevel<7 | ||
|---|---|---|---|
| Product: | Gentoo Linux | Reporter: | Christopher Head <bugs> |
| Component: | Current packages | Assignee: | Gentoo Kernel Bug Wranglers and Kernel Maintainers <kernel> |
| Status: | UNCONFIRMED --- | ||
| Severity: | normal | CC: | bugs |
| Priority: | Normal | ||
| Version: | unspecified | ||
| Hardware: | AMD64 | ||
| OS: | Linux | ||
| Whiteboard: | |||
| Package list: | Runtime testing required: | --- | |
| Attachments: |
Kernel config file with default log level 7
GRUB configuration file Bootconfig file Dmesg output from a successful boot |
||
|
Description
Christopher Head
2025-02-25 05:42:57 UTC
Anything on the screen ? With loglevel 6, there are a few console messages, but nothing obviously relevant: Booting a command line [ 0.000000] Linux version 6.12.16-gentoo (root@<hostname elided>) (gcc (Gentoo 14.2.1_p20241221 p7) 14.2.1 20241221, GNU ld (Gentoo 2.43 p3) 2.43.1) #4 Mon Feb 24 21:46:34 PST 2025 [ 0.011275] Kernel command line: BOOT_IMAGE=/root/boot/vmlinuz bootconfig loglevel=6 -- quiet [ 0.011305] Unknown kernel command line parameters "BOOT_IMAGE=/root/boot/vmlinuz", will be passed to user space. [ 0.011319] random: crng init done Please attach your full grub.cfg Created attachment 919948 [details]
GRUB configuration file
Sure, though there’s not much interesting about it.
Created attachment 919949 [details]
Bootconfig file
And before you ask, here’s the bootconfig.
Can I see a full dmesg of a working kernel? I've never seen a grub.cfg file like that. You make it with grub-mkconfig ? Another maybe dumb question, your boot partition if off /root ? Also, can you test with vanilla-sources and see if the same issue happens ? Created attachment 920031 [details]
Dmesg output from a successful boot
Here’s the dmesg output.
(In reply to Mike Pagano from comment #6) > I've never seen a grub.cfg file like that. You make it with grub-mkconfig ? No, hand-written. Given how simple it is to write a grub.cfg, I never quite found it worth the time to learn how the automation stuff works. > Another maybe dumb question, your boot partition if off /root ? No separate boot partition. There is a BIOS Boot Partition (not mounted, not a filesystem, just for GRUB to use, needed because it’s a GPT disk), a btrfs partition (which is root, and /boot is inside it), and a swap partition. > can you test with vanilla-sources and see if the same issue happens ? I tried vanilla-sources-6.12.16 (since that’s the same version number as the current stable gentoo-sources). Behaviour is exactly the same: hangs with loglevel=6, boots fine with loglevel=7. By the way, in case you’re wondering how I concluded this is the kernel hanging and not something else: 1. On this machine, if I boot with “debug loglevel=7”, I get a lot of kernel output, then the initramfs banner (“>>> Genkernel 4.3.16” and so on), and then the rescue shell. 2. On any other machine, if I boot with “debug loglevel=6”, I get short-ish kernel output but still the initramfs banner and rescue shell. 3. On this machine, if I boot with “debug loglevel=6”, I get only the very short kernel output I pasted above, and no initramfs banner. So by (1) I conclude that the initramfs on this machine is working fine if we can just get to it, by (2) I conclude that even with loglevel=6 the banner should still be printed if the get to the initramfs at all, and therefore by (3) we must not be getting to the initramfs. And it can’t be anything *before* the kernel hanging because those few kernel messages *are* printed. So if it’s not hanging pre-kernel, and it’s not hanging after reaching the initramfs, the kernel is the only thing left. |
