From: Henning Schild <henning.schild@siemens.com>
To: Jan Kiszka <jan.kiszka@siemens.com>
Cc: <isar-users@googlegroups.com>,
Florian Bezdeka <florian.bezdeka@siemens.com>,
Vijai Kumar K <Vijaikumar_Kanagarajan@mentor.com>
Subject: Re: [PATCH v2 0/7] re-fork wic pcbios and efi plugins
Date: Mon, 6 Sep 2021 10:59:02 +0200 [thread overview]
Message-ID: <20210906105902.35dcb338@md1za8fc.ad001.siemens.net> (raw)
In-Reply-To: <02091f27-08ca-a3ab-504e-ff1e15bff2cd@siemens.com>
Am Mon, 6 Sep 2021 07:05:17 +0200
schrieb Jan Kiszka <jan.kiszka@siemens.com>:
> On 03.09.21 14:53, Henning Schild wrote:
> > changes since v1:
> > - efi plugin forked as well
> > - systemd-boot support in efi plugin enabled
> > - common functionality in utility library
> > - test case for system-boot
> > - "cp -a" moved to "find exec cp" because of ubuntu
> > - changed wks files to exclude boot from root and mount it
> >
> > The forked plugins have gotten out of sync with the last wic version
> > bumps. And the original fork was not exactly minimal or made for
> > easy maintenance.
> >
> > This series does a re-fork of the two plugins with the aim to come
> > up with something readable, minimal and maintainable.
> >
>
> I already pointed this out to Baurzhan: We should create some
> developer doc somewhere that list the update procedures for forked
> elements. Which files come from upstream? Which are supposed to be
> vanilla? Which contain changes and should be revisited when upstream
> is updated? With such a file, both maintainers and contributors can
> check what to do when the want to update affected files / components
> or see patches touching them.
>
> Maybe start that file with covering wic and its plugins?
This is a good idea but i would not want to include a patch into this
series. Something to do on top.
The main reason is that we also have half-forked bits like the wks
files and especially meta/classes/wic-img.bbclass which probably has
not been checked in a long time.
> > There used to be a special case for grub-efi where the actual
> > kernel and initrd would remain in the root partition, which kind of
> > allowed kernel updates with apt-get. Now all three bootloaders
> > (systemd-boot now works as well) place bootloader, config and boot
> > artifacts in a boot-partition.
> >
> > Kernel updates with apt-get are now consistantly "broken". That
> > consistency very likely is not too bad. A generic solution for this
> > feature (if wanted) will need to be found. Covering not just these
> > three bootloaders but possibly also u-boot and efibootguard.
>
> I assume that this only applies to efi and pcbios images, not
> u-boot-script or anything else that manages kernel updates more or
> less properly already.
Right. The u-boot plugin is not a fork but our own. It might contain
copied bits but i did not look at it at all. And it is currently the
only bootloader bringing an update-script to generate bootloader conf
when updating the kernel with apt-get.
At a later point we can see if we can enable "apt-get update <kernel>"
bootloader indep. I think that should be possible very similar to how
we do things for u-boot.
But that is a story for the future, after this series has been merged.
I see it as a win to move from "mostly broken" to "consistently
broken". The only case that sometimes worked was grub ... depending on
which wic version and whether one would mount <boot> at /boot.
Henning
> Jan
>
next prev parent reply other threads:[~2021-09-06 8:59 UTC|newest]
Thread overview: 16+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-09-03 12:53 Henning Schild
2021-09-03 12:53 ` [PATCH v2 1/7] wic: reset our plugin forks to OE upstream for re-forking Henning Schild
2021-09-03 12:53 ` [PATCH v2 2/7] wic: add utility library for common bits of isar plugins Henning Schild
2021-09-03 12:53 ` [PATCH v2 3/7] wic: apply the actual fork changes to our pcbios plugin fork Henning Schild
2021-09-03 12:53 ` [PATCH v2 4/7] wic: clean up wic class in terms of isar variables Henning Schild
2021-09-03 12:53 ` [PATCH v2 5/7] wic: apply the actual fork changes to our efi plugin fork Henning Schild
2021-09-03 12:53 ` [PATCH v2 6/7] wic: mount /boot and exlude it from root for efi Henning Schild
2021-09-03 16:06 ` Henning Schild
2021-09-03 12:53 ` [PATCH v2 7/7] meta-isar: use "systemd-boot" for one test target Henning Schild
2021-09-06 5:05 ` [PATCH v2 0/7] re-fork wic pcbios and efi plugins Jan Kiszka
2021-09-06 8:59 ` Henning Schild [this message]
2021-09-06 9:48 ` Anton Mikanovich
2021-09-06 10:51 ` Henning Schild
2021-09-07 7:15 ` Bezdeka, Florian
2021-09-07 8:00 ` Henning Schild
2021-09-13 13:05 ` Anton Mikanovich
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=20210906105902.35dcb338@md1za8fc.ad001.siemens.net \
--to=henning.schild@siemens.com \
--cc=Vijaikumar_Kanagarajan@mentor.com \
--cc=florian.bezdeka@siemens.com \
--cc=isar-users@googlegroups.com \
--cc=jan.kiszka@siemens.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox