[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: What does it mean when GRUB fails to find a kernel symbol?
From: |
José Luiz Santana Dias |
Subject: |
Re: What does it mean when GRUB fails to find a kernel symbol? |
Date: |
Sun, 10 Sep 2023 14:47:29 +0000 |
I don't know to make this configuration. I have little knowlegde in Linux. No
understanding. I'm brazilian.
Obter o Outlook para Android<https://aka.ms/AAb9ysg>
________________________________
From: help-grub-bounces+jluizsantanadias=outlook.com@gnu.org
<help-grub-bounces+jluizsantanadias=outlook.com@gnu.org> on behalf of Saj
Goonatilleke <saj@discourse.org>
Sent: Saturday, September 9, 2023 4:59:56 PM
To: help-grub@gnu.org <help-grub@gnu.org>
Subject: Re: What does it mean when GRUB fails to find a kernel symbol?
Thanks, Pascal and Goh, for your replies.
(Your messages did not land in my mailbox, hence this self-reply.)
> Because the old core image lacks a symbol required by the new module.
I had assumed the symbol had been present in GRUB for a long time.
This was wrong. grub_disk_native_sectors was renamed from
grub_disk_get_size somewhat recently.
I think this rename provides half the answer.
The other half probably has something to do with the storied history of
these machines. They were updated in-place from a previous Debian major
release. Something may have gone wrong there. Perhaps we somehow ended
up with an old latent core image on one of our disks -- a core that was
built from before the symbol rename.
Anyway, yes, it seems this was another case of a core-module divergence.