Have been running 15.0 on a small profile Dell Optiplex 980 for some time. It overheats and shuts down in an ambient temperature of 74F-80F, so I bought a refurb 980 with a conventional size box and installed 15.0 on it using the same installation stick my son had made for me to install it on the small box. I then successfully booted up with the Huge kernel and ran Tabs Window Manager.

Next, I downloaded kernel-generic-5.15.145-i586-1.txz from a University of Utah mirror and used upgradepkg to change /boot/*5.15.19* --> /boot/*5.15.145*:

# upgradepkg kernel-generic-5.15.19-x86_64-2%kernel-generic-5.15.145-i586-1.txz

Then issued the following commands per the recipe found at https://docs.slackware.com/howtos:sl...e_for_generic:

# $(/usr/share/mkinitrd/mkinitrd_command_generator.sh -r)

# /usr/share/mkinitrd/mkinitrd_command_generator.sh -l /boot/vmlinuz-generic-5.15.145

# /usr/share/mkinitrd/mkinitrd_command_generator.sh -l /boot/vmlinuz-generic-5.15.145 >> /etc/lilo.conf

# cd /sbin
# ./lilo

I rebooted, selecting the generic kernel, then got a complaint at a point that the program needed to access /lib/modules/5.15.145

Not knowing a lot about what I was doing, I then downloaded linux-5.15.145/kernel-modules-5.15.145-i586-1.txz from the same mirror, and

# mkdir /lib/modules/5.15.145
# tar -Jxvf kernel-modules-5.15.145-i586-1.txz -C /lib/modules/5.15.145

Rebooted, selected generic kernel, and it hung up. Rebooted, selected huge kernel, and it hung up, I think in the same place, with the same output as for the generic:

[ 25.264933] Run /etc/init as init process
[ nn.nnnnnn] Run /bin/init as init process
[ nn.nnnnnn] Run /bin/sh as init process
[ nn.nnnnnn] request_module: kmod_concurrent_max (0) close to 0 (max_modprobes: 50), for module binfmt-464c, throttling...
[ nn.nnnnnn] request_module: modprobe binfmt-464c cannot be processed, kmod busy with 50 threads for than 5 seconds now
[ nn.nnnnnn] Starting init: /bin/sh exists but couldn't execute it (error -8)
[ nn.nnnnnn] Kernel panic - not syncing: no working init found. Try passing init = option 2 kernel. See Linux Documentation/admin-guide/init.rst for guidance.
[ nn.nnnnnn] CPU: 0 PID: 1 COMM: swapper Not tainted 5.15.145 #1
[ nn.nnnnnn] Hardware name: Dell Inc. OptiPlex 980 /0D441T, BIOS A05 11/03/2010
[ nn.nnnnnn] Call Trace:
[ nn.nnnnnn] ? panic+0x9e/0x224
[ nn.nnnnnn] ? rest_init+0x80/0x80
[ nn.nnnnnn] ? kernel_init+0xf2/0x100
[ nn.nnnnnn] ? ret_from_fork+0x1c/0x30
[ nn.nnnnnn] Kernel Offset: disabled
[ 30.385557] ---[ end Kernel panic - not syncing: No working init found. Try passing init= option to kernel. See Linux Documentation/admin-guide/init.rst for guidance. ]---

I can always just start over and use 5.15.19 but I am curious about what I have done wrong and/or omitted.

i586 means 32 bits. You "upgraded" your 64-bit system to run on a 32-bit kernel.

1 members found this post helpful.

So use the find command to first locate the modules for *5.15.145 then move them to /lib/modules/

So where is the initrd file?

Once you find the initrd file make sure that it is pointed to in the relevant part of your lilo.conf file.

Under you may be interested, I upgraded my kernel to 5.15.160 two days ago.So 5.15.145 is not the current kernel for 15 stable.

And under just curious is the computer running elilo or lilo?

A 32-bit kernel can't execute the 64-bit /sbin/init.

I marked this as solved but the reader should know that I was omitting something else, i.e., I had not installed the new version modules, only placed them in a directory. I am not planning on completing this upgrade soon, so I will not try to give more detail.

Yes, I am running lilo. On one of the last BIOS machines, I guess.

No more detail is needed. It's certain that kernel-generic-5.15.145-i586-1.txz can't work. You need kernel-generic-5.15.145-x86_64-1.txz.

1 members found this post helpful.