:: Re: [DNG] initramfs - what about fi…
Top Page
Delete this message
Reply to this message
Author: ael
Date:  
To: dng
Old-Topics: Re: [DNG] usr-merge
Subject: Re: [DNG] initramfs - what about firmware?
On Tue, Nov 18, 2025 at 11:10:20AM +0000, g4sra via Dng wrote:
>
>
> On Tuesday, November 18th, 2025 at 04:04, Steve Litt <slitt@???> wrote:
> -- snip --
> I DO.
> What is this obsession everybody has for complicating setups and increasing brittleness. There is only *two* good reasons for having an initramfs.
>
> 1) To get the root filesystem mounted - because you have some exotic root filesystem which is not supported by the kernel.
>
> 2) Because the initramfs *is* the root filesystem.


Is there not a 3rd and very important reason? I ask for information and
enlightenment.

When I have disassembled initramfs, the first stage is to *patch* the CPU
firmware to fix broken instructions and secuority problems. Only then
can you continue with a fully functioning and hopefully secure CPU.

I would much prefer to avoid initramfs especially when compiling a
custom kernel, but then I have to somehow patch the CPU firmware and then
essentially reboot into a renewed environment. Is there a standard way
to do this without an initramfs or a close equivalent?

I posted a similar question on this list a long time ago, and there was
no response which was worrying. Perhaps I am just showing my ignorance?

ael