Re: Question on ostree-init

[Paeglis Gatis <Gatis Paeglis theqtcompany com>]

> That said I would also take a patch to add a build-time option to ostree
to generate a statically linked ostree-prepare-root.

+1. Sounds very interesting for embedded.

________________________________________
From: ostree-list <ostree-list-bounces gnome org> on behalf of Colin Walters <walters verbum org>
Sent: Sunday, October 25, 2015 4:17 PM
To: ostree-list gnome org
Subject: Re: Question on ostree-init

On Fri, Oct 23, 2015, at 06:11 PM, Petros Aggelatos wrote:

> However, I afterwards discovered that this was how OSTree worked
> initially and the now abandoned ostree-init[1] project. So my question
> is, what lead to that decision and what are the disadvantages of the
> static init process?

Basically if you have an initramfs, there is little point to doing the
statically linked approach.  All my uses of ostree right now are for
"general purpose" OSes with an initramfs.

Now I know embedded-type approaches often want to avoid the boot-time
overhead of an initramfs.

A statically linked binary would work, but I think at that point it might be
better to put it in the kernel.  It'd be a fairly small patch I think; ostree-prepare-root.c
is only 252 lines of code with comments.

That said I would also take a patch to add a build-time option to ostree
to generate a statically linked ostree-prepare-root.
_______________________________________________
ostree-list mailing list
ostree-list gnome org
https://mail.gnome.org/mailman/listinfo/ostree-list