Flashes (Alerts)
Abstract
A KVM Guest in an LPAR can enter a lockup state after it is restored from an existing snapshot. This issue is seen when the snapshot is restored while a compute or memory intensive workload like stress-ng is running on the guest.
Content
Linux Release Affected
SLES-16 PPC64LE Kernels 6.12.0-160000.16-default or later
IBM Systems Affected
Any IBM Power System LPAR supporting and running a KVM Guest in an LPAR
Symptoms
After you restore a snapshot of a KVM Guest, the KVM guest can enter a lockup state. In such cases, the KVM guest displays the following kernel messages:
.361968] [ C45] CPU: 45 UID: 0 PID: 13520 Comm: stress-ng-vm Kdump: loaded Tainted: G L 6.12.0-160000.18-default #1 VOLUNTARY SLFO-1.2 (unreleased) d8b7565b2108528291f77a3ad716a06a353a0b84
[ 4261.361977] [ C45] Tainted: [L]=SOFTLOCKUP
[ 4261.361978] [ C45] Hardware name: IBM pSeries (emulated by qemu) Power11 (architected) 0x820200 0xf000007 of:SLOF,HEAD hv:linux,kvm pSeries
[ 4261.361981] [ C45] NIP: c00000000036bcd4 LR: c00000000036bc90 CTR: c000000000183640
[ 4261.361984] [ C45] REGS: c00000007428f250 TRAP: 0900 Tainted: G L (6.12.0-160000.18-default)
[ 4261.361988] [ C45] MSR: 8000000000009033 <SF,EE,ME,IR,DR,RI,LE> CR: 44042880 XER: 200400cd
[ 4261.361999] [ C45] CFAR: 0000000000000000 IRQMASK: 0
GPR00: c00000000036bc40 c00000007428f4f0 c000000001cf8100 000000000000000e
GPR04: 0000000000000040 000000000000000e 0000000000000000 000000000000000e
GPR08: c000000002b6a5e8 0000000000000001 c000000efbb19d40 c000000007a40e08
GPR12: c000000000183640 c000000effbd2300 00007fff66e90000 c00000007428f900
GPR16: 0000000000000001 000000000000002d fffffffffffffe7f c000000007a40780
GPR20: c000000002b7aa88 0000000000000000 c000000000160a70 0000000000000010
GPR24: c000000007a40780 0000000000000069 c000000000161d90 c000000002b6a5e8
GPR28: 0000000000000000 0000000000000178 c000000efd255400 c000000002b7c080
[ 4261.362044] [ C45] NIP [c00000000036bcd4] smp_call_function_many_cond+0x3b4/0x680
[ 4261.362054] [ C45] LR [c00000000036bc90] smp_call_function_many_cond+0x370/0x680
[ 4261.362060] [ C45] Call Trace:
[ 4261.362062] [ C45] [c00000007428f4f0] [c00000000036bc40] smp_call_function_many_cond+0x320/0x680 (unreliable)
[ 4261.362070] [ C45] [c00000007428f5b0] [c000000000160a70] flush_type_needed+0x1c0/0x240
[ 4261.362078] [ C45] [c00000007428f5f0] [c000000000161960] __radix__flush_tlb_range_psize+0x90/0x480
[ 4261.362084] [ C45] [c00000007428f680] [c0000000005d2b08] unmap_page_range+0xb18/0x1b60
[ 4261.362090] [ C45] [c00000007428f840] [c0000000005d3d50] unmap_vmas+0xb0/0x220
[ 4261.362095] [ C45] [c00000007428f8e0] [c000000000606d64] vms_clear_ptes.part.0+0xd4/0x160
[ 4261.362101] [ C45] [c00000007428f9b0] [c000000000608634] vms_complete_munmap_vmas+0x194/0x260
[ 4261.362107] [ C45] [c00000007428f9f0] [c00000000060903c] do_vmi_align_munmap+0x19c/0x1e0
[ 4261.362113] [ C45] [c00000007428fb10] [c00000000060c6c8] __vm_munmap+0x178/0x210
[ 4261.362118] [ C45] [c00000007428fbc0] [c0000000005e5f18] sys_munmap+0x28/0x40
[ 4261.362124] [ C45] [c00000007428fbe0] [c00000000002da9c] system_call_exception+0x12c/0x240
[ 4261.362130] [ C45] [c00000007428fe50] [c00000000000cfdc] system_call_vectored_common+0x15c/0x2ec
[ 4261.362138] [ C45] ---- interrupt: 3000 at 0x7fffa57687a4
[ 4261.362160] [ C45] NIP: 00007fffa57687a4 LR: 00007fffa57687a4 CTR: 0000000000000000
[ 4261.362163] [ C45] REGS: c00000007428fe80 TRAP: 3000 Tainted: G L (6.12.0-160000.18-default)
Workaround
Restarting the KVM guest can prevent this issue.
Fix Outlook
SUSE mirrored bug: Not applicable
The fix for this issue will be included in a later release.
I/O device impacted
None
Was this topic helpful?
Document Information
Modified date:
10 November 2025
UID
ibm17247829