Debian Buster/Sid PPC64

[xeno74]

From my side yes. The MintPPC diskimage is from Roland.

Thank you. As I have still problems with my powermac, I installed Debian from your image and upgrade to current packages.

All without problems. Now I will install few things like lxde, make backup and be ready for testing!

Great! Well done!

[xeno74]

I upgraded Debian 11 PPC64 to Debian Trixie/Sid PPC64 today.

First, I downloaded and installed the new keyring with dpkg -i.

New keyring: debian-ports-archive-keyring_2023.02.01_all.deb

After that, I updated the sources.list.

New /etc/apt/sources.list:

Code: Select all

# binary default
deb http://ftp.ports.debian.org/debian-ports/ unstable main
deb http://incoming.ports.debian.org/buildd/ unstable main
deb http://ftp.ports.debian.org/debian-ports/ unreleased main

# contrib and non-free arch:all packages (i.e. firmware)
deb [arch=all] http://ftp.debian.org/debian/ unstable contrib non-free-firmware

# source
deb-src http://ftp.debian.org/debian/ unstable main
deb-src http://incoming.debian.org/debian-buildd/ buildd-unstable main

Then I upgraded to Trixie with apt-get dist-upgrade.

I had an issue with usrmerge. (FATAL ERROR: Both /sbin/.... and /usr/sbin/.... exist)

I solved it with:

Code: Select all

apt --fix-broken install

Code: Select all

apt remove --purge usrmerge

Code: Select all

apt-mark hold usrmerge

Code: Select all

apt-mark hold usr-is-merged

[xeno74]

Currently, I can't upgrade Debian Sid PPC64:

Code: Select all

Setting up systemd (254-1) ...
systemd-machine-id-setup: error while loading shared libraries: libsystemd-shared-252.so: cannot open shared object file: No such file or directory
dpkg: error processing package systemd (--configure):
 installed systemd package post-installation script subprocess returned error exit status 127
Errors were encountered while processing:

The renaming of some systemd files has solved the issue:

Code: Select all

-rwxr-xr-x 1 root root  68008 Jun 11 23:05 /usr/bin/systemd-ask-password.bak
-rwxr-xr-x 1 root root  68152 Jun 11 23:05 /usr/bin/systemd-creds.bak
-rwxr-xr-x 1 root root  67888 Jun 11 23:05 /usr/bin/systemd-escape.bak
-rwxr-xr-x 1 root root  68112 Jun 11 23:05 /usr/bin/systemd-firstboot.bak
-rwxr-xr-x 1 root root  67912 Jun 11 23:05 /usr/bin/systemd-inhibit.bak
-rwxr-xr-x 1 root root  68008 Jun 11 23:05 /usr/bin/systemd-machine-id-setup.bak
-rwxr-xr-x 1 root root  67896 Jun 11 23:05 /usr/bin/systemd-notify.bak
-rwxr-xr-x 1 root root 199328 Jun 11 23:05 /usr/bin/systemd-repart.bak
-rwxr-xr-x 1 root root  68016 Jun 11 23:05 /usr/bin/systemd-sysext.bak
-rwxr-xr-x 1 root root 133792 Jun 11 23:05 /usr/bin/systemd-sysusers.bak
-rwxr-xr-x 1 root root 133704 Jun 11 23:05 /usr/bin/systemd-tmpfiles.bak
-rwxr-xr-x 1 root root  67888 Jun 11 23:05 /usr/bin/systemd-tty-ask-password-agent.bak

This issue is because of the directory moving to /usr (usrmerge).

/bin → /usr/bin
/sbin → /usr/sbin
/lib → /usr/lib
/lib64 → /usr/lib64

I don't have merged the bin, sbin, lib, and lib64 directories to /usr so I had some issues.

[xeno74]

Installation of packages via snapshots for example SuperTuxKart:

snapshot.debian.org

[xeno74]

How to set or change timezone on Debian:

Code: Select all

amigaone@debian-ppc64:~$ timedatectl set-timezone Europe/Berlin
amigaone@debian-ppc64:~$ timedatectl
               Local time: Tue 2023-08-15 11:04:22 CEST
           Universal time: Tue 2023-08-15 09:04:22 UTC
                 RTC time: Tue 2023-08-15 09:04:22
                Time zone: Europe/Berlin (CEST, +0200)
System clock synchronized: yes
              NTP service: active
          RTC in local TZ: no

[Hypex]

My PPC64LE test:

Code: Select all

[02:02:25:209] ␍Booting configuration PPC64LE␊
[02:02:25:229] ␍Loaded modules in 7.055 sec␊
[02:02:32:312] ␍<0x1b>[2J<0x1b>[31;1mX1Boot Loader by Damien Stewart.␍␊
[02:02:32:414] <0x1b>[0m<0x1b>[35m␍␊
[02:02:32:435] ␍␊
[02:02:32:435] Kernel:  vmlinux-6.4␍␊
[02:02:32:435] <0x1b>[34mUnpacking XZ␍␊
[02:02:32:435] <0x1b>[35m␍␊
[02:02:35:873] Ramdisk: initrd-6.4␍␊
[02:02:35:873] <0x1b>[34mUnpacking XZ␍␊
[02:02:35:873] <0x1b>[35m␍␊
[02:02:36:322] Options: root=/dev/ram console=ttyS0,115200 console=tty0<0x1b>[34m␍␊
[02:02:36:334] Set␍␊
[02:02:36:334] ␍␊
[02:02:36:334] <0x1b>[32m␍␊
[02:02:36:356] Entering kernel...␍␊
[02:02:36:356] <0x1b>[37m␍␊
[02:02:36:356] ␍␊
[02:02:36:356] OF stdout device is: /bootconsole␍␊
[02:02:36:366] Preparing to boot Linux version 6.4.0-2-powerpc64le ([email protected]) (gcc-13 (Debian 13.2.0-1) 13.2.0, GNU ld (GNU Binutils for Debian) 2.41) #1 SMP Debian 6.4.4-3 (2023-08-08)␍␊
[02:02:36:383] Detected machine type: 0000000000000500␍␊

That's as far as it gets. Nothing on serial. Wonder if something is missing from my ELF loader but it's only a test.

[Hypex]

So last night, or overnight really, I ran the Debian ppc64 installer. The latest Debian GNU/Linux 12.0.0 "Sid" - Unofficial ppc64 NETINST 20240225-20:36. I used the 6.12.17 kernel with installer ramdisk. I wrote the CD image to a USB stick and the installer picked it up as a cdrom source. This was useful as it means you don't need to burn a CD and can simply write the image to a USB stick like a modern person.

Unfortunately, being unstable, it really is unstable. Main binaries like apt just crash. A few others do as well. It would be good if a stable release could be specified that just pulls in known stable components. Would be funny if it was that easy.

I left it overnight pulling in desktop components. The installer apt works fine. But the installed target apt is broken. This means that running task select will always fail, because it runs it on the target, which will crash. There's errors with '/var/lib/dpkg/status' from the start missing fields. It tries to run systemctl but it's not installed. The used debootstrap has an 'if' bug and needs to be shellchecked. Checking the errors it's almost funny. But it is on unofficial ppc64 so even a generic shell script can be broken it looks, though for such a generic file, I thought it would be tested by the build process and red flag it. I should probably send the details to the ppc64 team.

[xeno74]

Latest sources.list (/etc/apt/sources.list):

Code: Select all

# binary default
deb http://ftp.ports.debian.org/debian-ports/ unstable main
deb http://incoming.ports.debian.org/buildd/ unstable main
deb http://ftp.ports.debian.org/debian-ports/ unreleased main

# contrib and non-free arch:all packages (i.e. firmware)
deb [arch=all] http://ftp.debian.org/debian/ unstable contrib non-free non-free-firmware

# source
deb-src http://ftp.debian.org/debian/ unstable main
deb-src http://incoming.debian.org/debian-buildd/ buildd-unstable main

[Hypex]

I've got a bit further with Debian Sid. So I disabled GLX in the xorg config. This allowed Epiphany to load without taking down the system. But it still didn't work. It kept giving errors like being unable to render page and problem loading page. I didn't expect it but it seems that Epiphany needs hardware 3d drivers in order to work at all.

In any case the system log is filled with the GPU lockup errors. Which are known in the x86 world. Some point to hardware failure. But my card works fine on all other OS installs. Currently the system log has relaxed with all the constant errors so I can reboot quicker. But the system is still unusable for any practical use.

[Hypex]

I've got SeaLion working on my Debian 64!

I needed the libffi6 package. I didn't need the libjpeg8 package but I located it just in case. I've found a snapshot for both PPC32 and PPC64.

FFI6:
https://snapshot.debian.org/archive/deb ... _ppc64.deb
https://snapshot.debian.org/archive/deb ... owerpc.deb

JPEG8:
https://snapshot.debian.org/archive/deb ... _ppc64.deb
https://snapshot.debian.org/archive/deb ... owerpc.deb