Flashes (Alerts)
Abstract
A KVM guest that is running in an LPAR might enter a lockup state after multiple migration attempts under heavy system load. This issue typically occurs when consecutive migrations are performed in quick succession while a compute- or memory-intensive workload, such as 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 supports running a KVM Guest in an LPAR.
Symptoms
Performing multiple migration operations on a KVM Guest running with memory or cpu intensive workloads (for example, stress-ng) can result in the KVM guest entering a lockup state after the migration operation completes. In such cases, the KVM guest shows the following kernel messages:
[12584.019115] [ C11] NIP [c00000000036cac4] smp_call_function_many_cond+0x3b4/0x680
[12584.019120] [ C11] LR [c00000000036ca80] smp_call_function_many_cond+0x370/0x680
[12584.019123] [ C11] Call Trace:
[12584.019123] [ C11] [c00000001907f8d0] [c00000000036cd44] smp_call_function_many_cond+0x634/0x680 (unreliable)
[12584.019127] [ C11] [c00000001907f990] [c000000000160a70] flush_type_needed+0x1c0/0x240
[12584.019131] [ C11] [c00000001907f9d0] [c000000000161960] __radix__flush_tlb_range_psize+0x90/0x480
[12584.019134] [ C11] [c00000001907fa60] [c0000000005eb2cc] tlb_finish_mmu+0x18c/0x1f0
[12584.019138] [ C11] [c00000001907faa0] [c000000000607c18] vms_clear_ptes.part.0+0x118/0x160
[12584.019141] [ C11] [c00000001907fb70] [c0000000006094c4] vms_complete_munmap_vmas+0x194/0x260
[12584.019144] [ C11] [c00000001907fbb0] [c000000000609ecc] do_vmi_align_munmap+0x19c/0x1e0
[12584.019147] [ C11] [c00000001907fcd0] [c00000000060d578] __vm_munmap+0x178/0x210
[12584.019150] [ C11] [c00000001907fd80] [c0000000005e6d78] sys_munmap+0x28/0x40
[12584.019152] [ C11] [c00000001907fda0] [c00000000002da9c] system_call_exception+0x12c/0x240
[12584.019155] [ C11] [c00000001907fe50] [c00000000000d620] system_call_common+0x160/0x2c4
[12584.019159] [ C11] ---- interrupt: c00 at 0x7fffb9901724
[12584.019162] [ C11] NIP: 00007fffb9901724 LR: 00007fffb98c3cb8 CTR: 0000000000000000
[12584.019164] [ C11] REGS: c00000001907fe80 TRAP: 0c00 Tainted: G L (6.12.0-160000.5-default)
[12584.019165] [ C11] MSR: 800000000280f033 <SF,VEC,VSX,EE,PR,FP,ME,IR,DR,RI,LE> CR: 24002824 XER: 00000000
[12584.019171] [ C11] IRQMASK: 0
GPR00: 000000000000005b 00007fffeef1d130 00007fffb9937f00 00007fffb8700000
GPR04: 0000000000020010 0000000000000001 000000000000d298 0000000000000000
GPR08: 0000000000000000 0000000000000000 0000000000000000 0000000000000000
GPR12: 0000000000000000 00007fffb9943b00 0000000000000000 00007fffeef1d2a0
GPR16: 00007fffb9930000 00007fffb9930000 0000000000000000 0000000000000000
GPR20: ffffffffffffffff ffffffffffffffff 0000000000000000 00007fffb9930000
GPR24: 00007fffeef1d190 0000000000000001 00007fffeef1d2f8 ffffffffffffffff
GPR28: 00000001181db4f0 0000000000000000 0000000000000000 00007fffeef1d2a0
[12584.019190] [ C11] NIP [00007fffb9901724] 0x7fffb9901724
[12584.019191] [ C11] LR [00007fffb98c3cb8] 0x7fffb98c3cb8
[12584.019193] [ C11] ---- interrupt: c00
[12584.019193] [ C11] Code: e95e0000 390924e8 7c691ef4 7d28482a 7d4a4a14 812a0008 71290001 41820024 60000000 60000000 60000000 7c40003c <60000000> 812a0008 71290001 4082fff0
[12604.009035] [ C8] watchdog: BUG: soft lockup - CPU#8 stuck for 2777s! [sleep:106423]Workaround
Avoid performing multiple migrations in quick succession on a KVM guest that runs compute or memory intensive workloads.
Fix Outlook
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
ibm17247968