From: Morten Linderud <morten@linderud.pw>
To: "Philip Müller" <philm@manjaro.org>
Cc: Javier Martinez Canillas <javierm@redhat.com>,
The development of GNU GRUB <grub-devel@gnu.org>,
Mike Gilbert <floppym@gentoo.org>,
Daniel Kiper <daniel.kiper@oracle.com>,
Bernhard Landauer <bernhard@manjaro.org>,
Mark Wagie <mark@manjaro.org>,
Robbie Harwood <rharwood@redhat.com>
Subject: Re: [Regression] efi: Don't display a uefi-firmware entry if it's not supported
Date: Wed, 31 Aug 2022 13:16:15 +0200 [thread overview]
Message-ID: <20220831111615.cg5wwdqbh2jrunc6@framework> (raw)
In-Reply-To: <617d8190-0999-9b5f-f809-c4fccf7572cc@manjaro.org>
On Wed, Aug 31, 2022 at 12:07:33AM +0200, Philip Müller wrote:
> On 30.08.22 23:34, Javier Martinez Canillas wrote:
> > > You could add a feature flag, which causes grub-core to set an
> > > environment variable when a new feature is supported. See the features
> > > array in grub-core/normal/main.c.
> > >
> > > You would then check for this feature flag in the grub.d snippet
> > > before calling fwsetup --is-supported.
> > >
> > Please don't. As mentioned, I think we should aim to simplify the grub.cfg
> > instead of making it more complicated.
>
> Well I think it would be the best approach to add backward compatibility as
> most users don't even know on how to install grub via grub-install. That is
> done via the graphical installer Calamares on most Arch-based Distributions.
> Updating the grub menu is common if you install multiple kernels or use
> snapshots via BTRFS.
Lets make a clear distinction between how Arch, the package upstream in this
case, and how derivative distros like Manjaro, the downstream user of this
package, uses grub though. The intent behind the packaging on our end does not
match up with what you the expectations you are conveying here.
Arch doesn't make assumptions about the users system and since `grub-install` is
a complicated command we can't automate this for users. If they use Secure Boot
they need to include modules, where is the ESP located and so on.
Arch requires the users to know how this command works, and it's a fair
assumption as it's documented in our install instructions. The fact that
derivative distros relies on this decision from Arch is their problem and should
not necessarily be a grub upstream issue.
Other distros build and ship monolithic grub.efi binaries for Secure Boot and
shim setups. From what I understand this entails that the configuration and the
binary stays in sync. This is not the case for Arch currently but might change
in the future.
If grub needs the configuration and grub.efi binaries to stay in sync, then it
would be much better for `grub-install` to support a config file like how people
have wound up using `grub-mkconfig`. Then shipping a hook that simply runs
`grub-install` would be simpler.
I think this is a good future solution as it would remove the need for distros
to ship `grub-install` wrappers anyway.
--
Morten Linderud
PGP: 9C02FF419FECBE16
next prev parent reply other threads:[~2022-08-31 13:36 UTC|newest]
Thread overview: 14+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <9274cc62-1922-76c8-925b-b172389b6c3d@manjaro.org>
[not found] ` <897c60aa-a254-09e3-9efa-a221cd58d2a9@manjaro.org>
2022-08-30 20:16 ` [Regression] efi: Don't display a uefi-firmware entry if it's not supported Robbie Harwood
2022-08-30 21:14 ` Mike Gilbert
2022-08-30 21:34 ` Javier Martinez Canillas
2022-08-30 22:07 ` Philip Müller
2022-08-30 22:43 ` Javier Martinez Canillas
2022-08-30 23:15 ` Philip Müller
2022-08-31 4:46 ` Luna Jernberg
2022-08-31 18:44 ` Robbie Harwood
2022-09-01 17:11 ` Philip Müller
2022-08-31 11:16 ` Morten Linderud [this message]
2022-08-31 11:47 ` Mihai Moldovan
2022-08-31 15:14 ` Dimitri John Ledkov
2022-09-05 22:49 ` Glenn Washburn
2022-10-28 12:55 ` Daniel Kiper
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20220831111615.cg5wwdqbh2jrunc6@framework \
--to=morten@linderud.pw \
--cc=bernhard@manjaro.org \
--cc=daniel.kiper@oracle.com \
--cc=floppym@gentoo.org \
--cc=grub-devel@gnu.org \
--cc=javierm@redhat.com \
--cc=mark@manjaro.org \
--cc=philm@manjaro.org \
--cc=rharwood@redhat.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.