PSOD with an error "Spin count exceeded - possible deadlock" with "vmk_PriQDestroy" in the backtrace

Products

• VMware vSphere ESXi

Issue/Introduction

• ESXi host fails with a PSOD as below:

VMware ESXi 8.0.x [Releasebuild-########x86_64]
Spin count exceeded - possible deadlock

Module(s) involved in panic: [nmp Version 1.0.0-0, Built on: MMM DD YYYY]
cr0=0x8001003d cr2=0x30 cr3=0x84ca6000 cr4=0x14216c
FMS=06/55/4 uCodc=0x2007006
*PCPU43:2097914/NMP Device Registration Helper
PCPU B: VSVVVVVVVSSSVVSSSVVVSVVVVVVVVVVVVVVVSVVSVVVSVVVVVVVVVVSVVVVVVVVV
PCPU 64: VSVVVV
Code start: 0x42000dc00000 VMK uptime: DD:HH:MM:SS
##############: [##############]PanicvPanicInt@vmkernel#nover+0x202 stack: 0xe
##############: [##############]Panic_NoSave@vmkernel#nover+0x4d stack: 0x45399379bc90
##############: [##############]Lock_CheckSpinCount@vmkernel#nover+0x269 stack: 0x430a41e00f63
##############: [##############]SP_HaitLockIRQ@vmkernel#nover+0xf0 stack: 0x0
##############: [##############]SPLockIRQWork@vmkernel#nover.0x5c stack: 0x43136c802c40
##############: [##############]Heap_Free@vmkernel#nover+0x62 stack: 0x430a41e00000
##############: [##############]vmk_PriQDestroy@vmkernel#nover+0x22 stack: 0x0
##############: [##############]StorSchedCleanupCommandQueueData@vmkernel#nover+0x1a stack: 0x4313692205c0
##############: [##############]ScsiAllocateDevice@vmkernel#nover+0x9b2 stack: 0x430a5a843ae8
##############: [##############]vmk_ScsiAllocateDevice@vmkernel#nover.0x40 stack: 0x43136e8b2740
##############: [##############][email protected]#v2_12_0_0+0x19 stack: 0x43136e8b2740
##############: [##############][email protected]#v2_12_0_0.0xcc stack: 0x45399379f000
##############: [##############][email protected]#v2_12_0_0+0x10 stack: 0x431332001720
##############: [##############]HelperQueueFunc@vmkernel#nover+0x300 stack: 0x431332001238
##############: [##############]CpuSched_StartWorld@vmkernel#nover+0xe2 stack: 0x0
##############: [##############]Debug_IsInitialized@vmkernel#nover+0xc stack: 0x0
base fs=0x0 gs=0x42004ac00000 Kgs=0x0
YYYY-MM-DDTHH:MM cpu43:2097914)NMI: 743: NMI IPI: RIPOFF(base) :RBP:CS [0x124299(0x42000dc00000):0x430a41e00f60:0x748] (Src 0x1, CPU43)
YYYY-MM-DDTHH:MM cpu43:2097914)NMI: 743: NMI IPI: RIPOFF(base):RBP:CS [0x124276(0x42000dc00000):0x430a41e00f60:0x748] (Src 0x1, CPU43)
YYYY-MM-DDTHH:MM cpu48:2097724)Jumpstart plugin restore-security-policies activation failed.
YYYY-MM-DDTHH:MM cpu45:2097724)Jumpstart plugin restore-networking activation failed.
No place on disk to dump data.
Finalized dump header (16/16) FileDump: Successful.
No port for remote debugger.

Environment

• VMware vSphere ESXi 8.x

Cause

• Load unload of driver module on the host with high number of ports.

Resolution

The issue is fixed in VMware ESXi 8.0 Update 3b. Log in to the Broadcom Support Portal to download this patch.