On 03/01/2023 18:55, Vagrant Cascadian wrote:
On 2023-01-03, Frédéric Danis wrote:
On 29/12/2022 00:07, Vagrant Cascadian wrote:
On 2022-12-28, Vagrant Cascadian wrote:
The odroid-c2 fails to boot syslinux/extlinux style menus (e.g. those
produced by u-boot-menu) or boot.scr as of upstream 2022.07-rc1. The
commit triggering the issue has been identified as:
a9bf024b2933bba0e23038892970a18b72dfaeb4
efi_loader: disk: a helper function to create efi_disk objects from
udevice
Workarounds I've heard are to disable EFI support for that board, or to
boot using EFI rather than boot scripts or syslinux/extlinux style
menus.
I will also want to get confirmation if other amlogic boards are
affected...
The currently supported amlogic platforms are:
# Neil Armstrong <narmstr...@baylibre.com>
u-boot-amlogic_platforms += khadas-vim
# Neil Armstrong <narmstr...@baylibre.com>
u-boot-amlogic_platforms += khadas-vim2
# Frederic Danis <frederic.da...@collabora.com>
u-boot-amlogic_platforms += libretech-cc
# Neil Armstrong <narmstr...@baylibre.com>
u-boot-amlogic_platforms += nanopi-k2
# Vagrant Cascadian <vagr...@debian.org>
u-boot-amlogic_platforms += odroid-c2
# Reco <b...@enotuniq.net>
u-boot-amlogic_platforms += odroid-n2
Please test if the current versions from Debian unstable (2022.10*) and
experimental (2023.01-rc*) are affected by this issue... and if there
are other issues for Debian bookworm/testing (2022.04*).
This is part of what has been blocking u-boot from migrating to testing.
I do not see (m)any records of tests for most of these platforms at:
https://wiki.debian.org/U-boot/Status
live well,
vagrant
u-boot bookworm version works fine with LePotato board (libretech-cc),
see attached lepotato-bookworm.txt.
Thanks!
Afaiu, the bug is not reproducible with unstable and experimental
versions, but boot crash before starting the kernel, see
lepotato-unstable.txt and lepotato-experimental.txt.
Sounds like the bug *is* reproducible with unstable and experiemental
versions, from reading the logs; looks like the same issue with
odroid-c2.
Could you try building a "noefi" variant, like done with odroid-c2, and
test for the lepotato (libretech-cc?):
https://salsa.debian.org/debian/u-boot/-/commit/711ca985af1cab6f11e33fcf5e30dcc40dc7e64d
The "noefi" variant works fine for unstable (2022.10+dfsg-2) and
experimental (2023.01~rc4+dfsg-1) versions.
I create a merge request for it:
https://salsa.debian.org/debian/u-boot/-/merge_requests/28
Could you also test EFI booting with the versions from unstable and
experimental to see if they can load a kernel? If EFI does work, that
justifies building two variants with and without EFI... it is a bit ugly
to produce two variants, but it's the easiest known workaround for
now...
I'm able to boot the debian-installer mini.iso and get the first install
menu with unstable and experimental u-boot.
Regards,
Fred
--
Frédéric Danis
Senior Software Engineer
Collabora Ltd.
Platinum Building, St John's Innovation Park, Cambridge CB4 0DS, United Kingdom
Registered in England & Wales, no. 5513718