2

First of all, on my desktop running Ubuntu 22.04 with 6.5.0-14 kernel and VirtualBox 7.0.14 everything works fine. No error messages. But if I try the same combination on my laptop ThinkPad T450s I get the following message in syslog and hundreds of similar messages, one after another:

Jan 21 13:38:50 thinkpad kernel: [  150.108653] UBSAN: array-index-out-of-bounds in /tmp/vbox.0/r0drv/linux/memobj-r0drv-linux.c:1897:78
Jan 21 13:38:50 thinkpad kernel: [  150.108657] index 1 is out of range for type 'page *[1]'
Jan 21 13:38:50 thinkpad kernel: [  150.108660] CPU: 3 PID: 2060 Comm: VirtualBoxVM Tainted: G           OE      6.5.0-14-generic #14~22.04.1-Ubuntu
Jan 21 13:38:50 thinkpad kernel: [  150.108663] Hardware name: LENOVO 20BX000TUK/20BX000TUK, BIOS JBET49WW (1.14 ) 05/21/2015
Jan 21 13:38:50 thinkpad kernel: [  150.108665] Call Trace:
Jan 21 13:38:50 thinkpad kernel: [  150.108666]  <TASK>
Jan 21 13:38:50 thinkpad kernel: [  150.108669]  dump_stack_lvl+0x48/0x70
Jan 21 13:38:50 thinkpad kernel: [  150.108678]  dump_stack+0x10/0x20
Jan 21 13:38:50 thinkpad kernel: [  150.108680]  __ubsan_handle_out_of_bounds+0xc6/0x110
Jan 21 13:38:50 thinkpad kernel: [  150.108685]  rtR0MemObjNativeMapUser+0x3d3/0x3e0 [vboxdrv]
Jan 21 13:38:50 thinkpad kernel: [  150.108722]  VBoxHost_RTR0MemObjMapUserExTag+0x124/0x280 [vboxdrv]
Jan 21 13:38:50 thinkpad kernel: [  150.108757]  VBoxHost_RTR0MemObjMapUserTag+0x15/0x30 [vboxdrv]
Jan 21 13:38:50 thinkpad kernel: [  150.108791]  SUPR0GipMap+0x39d/0x5c0 [vboxdrv]
Jan 21 13:38:50 thinkpad kernel: [  150.108819]  supdrvIOCtl+0x137b/0x3620 [vboxdrv]
Jan 21 13:38:50 thinkpad kernel: [  150.108845]  ? _copy_from_user+0x30/0xa0
Jan 21 13:38:50 thinkpad kernel: [  150.108849]  VBoxDrvLinuxIOCtl_7_0_14+0x17c/0x270 [vboxdrv]
Jan 21 13:38:50 thinkpad kernel: [  150.108873]  __x64_sys_ioctl+0xa3/0xf0
Jan 21 13:38:50 thinkpad kernel: [  150.108877]  do_syscall_64+0x5b/0x90
Jan 21 13:38:50 thinkpad kernel: [  150.108880]  ? exit_to_user_mode_prepare+0x30/0xb0
Jan 21 13:38:50 thinkpad kernel: [  150.108884]  ? syscall_exit_to_user_mode+0x37/0x60
Jan 21 13:38:50 thinkpad kernel: [  150.108887]  ? do_syscall_64+0x67/0x90
Jan 21 13:38:50 thinkpad kernel: [  150.108889]  ? do_user_addr_fault+0x17a/0x6b0
Jan 21 13:38:50 thinkpad kernel: [  150.108891]  ? exit_to_user_mode_prepare+0x30/0xb0
Jan 21 13:38:50 thinkpad kernel: [  150.108894]  ? irqentry_exit_to_user_mode+0x17/0x20
Jan 21 13:38:50 thinkpad kernel: [  150.108897]  ? irqentry_exit+0x43/0x50
Jan 21 13:38:50 thinkpad kernel: [  150.108899]  ? exc_page_fault+0x94/0x1b0
Jan 21 13:38:50 thinkpad kernel: [  150.108902]  entry_SYSCALL_64_after_hwframe+0x6e/0xd8
Jan 21 13:38:50 thinkpad kernel: [  150.108905] RIP: 0033:0x7f0d0331a94f
Jan 21 13:38:50 thinkpad kernel: [  150.108925] Code: 00 48 89 44 24 18 31 c0 48 8d 44 24 60 c7 04 24 10 00 00 00 48 89 44 24 08 48 8d 44 24 20 48 89 44 24 10 b8 10 00 00 00 0f 05 <41> 89 c0 3d 00 f0 ff ff 77 1f 48 8b 44 24 18 64 48 2b 04 25 28 00
Jan 21 13:38:50 thinkpad kernel: [  150.108927] RSP: 002b:00007ffe9d1db4b0 EFLAGS: 00000246 ORIG_RAX: 0000000000000010
Jan 21 13:38:50 thinkpad kernel: [  150.108929] RAX: ffffffffffffffda RBX: 0000000000000000 RCX: 00007f0d0331a94f
Jan 21 13:38:50 thinkpad kernel: [  150.108931] RDX: 00007ffe9d1db550 RSI: 00000000c0305694 RDI: 0000000000000007
Jan 21 13:38:50 thinkpad kernel: [  150.108932] RBP: 00007ffe9d1db510 R08: 0000000000000000 R09: 000055d290e6a970
Jan 21 13:38:50 thinkpad kernel: [  150.108933] R10: 0000000000000077 R11: 0000000000000246 R12: 0000000000000000
Jan 21 13:38:50 thinkpad kernel: [  150.108934] R13: 00007f0d0314e508 R14: 00007f0d030e03f0 R15: 000055d290e6a970
Jan 21 13:38:50 thinkpad kernel: [  150.108936]  </TASK>
Jan 21 13:38:50 thinkpad kernel: [  150.108937] ================================================================================

Virtualbox 6 stopped working After kernel upgrade in Ubuntu 22.04

GAD3R
  • 3,507
Tigran Aivazian
  • 177
  • Correction: the same errors are generated on my desktop as well, I just didn't notice them originally. So now I have to downgrade the desktop back to 6.2.0-39 kernel as well... – Tigran Aivazian Jan 21 '24 at 15:43
  • 2
    They have a bug open on this over at VirtualBox https://www.virtualbox.org/ticket/21877. Unfortunately, no fixes yet. Looks like I missed this on mine, but just because mine is spamming these errors only when I run VirtualBox or /sbin/vboxconfig that it hits my syslog file with these yet, none of my VMs are failing in anyway shape or form. – Terrance Jan 21 '24 at 15:58
  • @Terrance Well, I get them on every boot, probably because /sbin/vboxconfig is executed on every boot. If you check carefully you will probably find you have those error message on every boot too. – Tigran Aivazian Jan 22 '24 at 08:28
  • While that is true that they do appear on boot, they are not causing any of my VMs to not work. And, also, not every VM I have causes this. Legacy VMs like DOS or Windows 98 do not cause the panic message, which is output only from UBSAN. So, it's just more of a nuisance than anything else and it is something that Oracle themselves will have to fix. Nothing we can do here to fix these messages, unless you want to disable UBSAN which is not recommended. – Terrance Jan 22 '24 at 14:05
  • @Terrance Computers are devices for storing and processing information and information is a very tricky thing. One bit somewhere can be flipped and you will never notice until you come to verify the MD5 checksum thereof, and that is only if have previously recorded the checksum, as I always do for important data. – Tigran Aivazian Jan 23 '24 at 17:15
  • Just to add that the problem persists also with kernel 6.5.0-15 and VirtualBox 7.0.15. – Tigran Aivazian Jan 30 '24 at 08:11

1 Answers1

1

The "solution" was to reboot back into 6.2.0-39 kernel, where everything worked fine.

Except that this solution works only on the laptop. On the desktop it does NOT work, because I have multiple monitors and the recent (a few days ago) upgrade in Ubuntu 22.04 broke something in the X11 support, so that even if I disable Xwayland in /etc/gdm3/custom.conf I still cannot boot with 6.2.0-39 and use more than one monitor and even that only in 1024x768 resolution instead of native 4K. So on the desktop I have to boot with 6.5.0-14 and just ignore the error messages in the system log. VMs seem to work fine.

Tigran Aivazian
  • 177