No VM starting with own Kernel beyond Version 6.14

[Herbert99]

Hi!

I had upgraded Ubuntu 25.04 to 25.10 with Kernel 6.14. to 6.17.
I had my own Kernel 6.14 config, which worked well with Vbox then and still works with Vbox 7.2.2-dfsg-2.
But my own new Kernel 6.17 does not work. Its config is based on my 6.14 config.
No machine could boot and crashed instantly with a kernel error message:

Code: Select all

EMT-0[12862]: segfault at 7f9a1d4ddff0 ip 00007f9a1d4ddff0 sp 00007f9a047f7c88 error 14 in VBoxVMM .so[2ddff0,7f9a1d2a3000+303000] likely on CPU 12 (core 4, socket 0)
kernel: Code: Unable to access opcode bytes at 0x7f9a1d4ddfc6.
kernel: BUG: scheduling while atomic: EMT-0/12862/0x7ffff801

Kernel 6.16 with config based on 6.14 does not work too.
Ubuntu stock kernel 6.17 works.

All kernels are booted with:
ro verbose mem_encrypt=off delayacct debugfs=on vsyscall=none zswap.enabled=1 zswap.zpool=zsmalloc slab_debug=-

There is no error in VM log, they all think, they are running. Last lines are:
00:00:01.016182 Changing the VM state from 'POWERING_ON' to 'RUNNING'
00:00:01.016200 Console: Machine state changed to 'Running'

The Vbox boot screen is not shown in any VM.

[Voncloft]

Same result, tried to use all versions of Virtualbox for 6.x and 7.x no dice

kernel version: 6.15.4

Code: Select all

[27160.474447] EMT[26360]: segfault at 7f5ee34fc180 ip 00007f5ee34fc180 sp 00007f5f0e2ceaf8 [48180,7f5ee34ea000+2e7000] likely on CPU 6 (core 2, socket 0)
[27160.474460] Code: Unable to access opcode bytes at 0x7f5ee34fc156.
[27160.474523] BUG: scheduling while atomic: EMT/26360/0x7fffce01
[27160.474526] Modules linked in: vboxnetflt(O) vboxnetadp(O) vboxdrv(O) nvidia_uvm(PO) lgdt3306a nvidia_drm(PO) nvidia_modeset(PO) nvidia(PO) xpad(O) x86_pkg_temp_thermal snd_hda_codec_realtek snd_hda_codec_generic snd_hda_codec_hdmi snd_hda_scodec_component snd_hda_intel snd_intel_dspcfg snd_hda_codec snd_hwdep snd_hda_core snd_pcm saa7164 snd_timer snd soundcore tveeprom [last unloaded: vboxdrv(O)]
[27160.474549] CPU: 6 UID: 1000 PID: 26360 Comm: EMT Tainted: P        W  O        6.15.4 #1 PREEMPT(voluntary) 
[27160.474554] Tainted: [P]=PROPRIETARY_MODULE, [W]=WARN, [O]=OOT_MODULE
[27160.474555] Hardware name: To Be Filled By O.E.M. To Be Filled By O.E.M./Z270 Professional Gaming i7, BIOS P1.10 12/29/2016
[27160.474557] Call Trace:
[27160.474559]  <TASK>
[27160.474561]  dump_stack_lvl+0x4d/0x70
[27160.474568]  __schedule_bug.cold+0x3e/0x4a
[27160.474571]  __schedule+0xb44/0xf20
[27160.474575]  ? __lruvec_stat_mod_folio+0xa1/0xd0
[27160.474579]  do_task_dead+0x46/0xa0
[27160.474583]  do_exit+0x7e7/0xac0
[27160.474586]  do_group_exit+0x28/0xb0
[27160.474589]  get_signal+0x8a0/0x8d0
[27160.474592]  arch_do_signal_or_restart+0x36/0x260
[27160.474598]  irqentry_exit_to_user_mode+0xb7/0x120
[27160.474602]  asm_exc_page_fault+0x26/0x30
[27160.474605] RIP: 0033:0x7f5ee34fc180
[27160.474607] Code: Unable to access opcode bytes at 0x7f5ee34fc156.
[27160.474608] RSP: 002b:00007f5f0e2ceaf8 EFLAGS: 00010206
[27160.474611] RAX: 000000000000000b RBX: 000000001c82be50 RCX: 000000000000000b
[27160.474613] RDX: 0000000000000060 RSI: 00007f5f0df77e80 RDI: 00007f5f0df76000
[27160.474614] RBP: 00007f5f0e2cec40 R08: 0000000000000001 R09: 0000000000000000
[27160.474616] R10: 0000000000000000 R11: 00007f5f0e2cf6c0 R12: 00007f5f0df76000
[27160.474618] R13: 00007f5ee37d4211 R14: 00007f5f0df77e80 R15: 00007f5f0df762c0
[27160.474620]  </TASK>

[Herbert99]

Did you some kernel hardening?

[Herbert99]

I had to load kvm modules.