Hello,
This NTDEV post makes me think the combination of driver verifier and a debug build of my driver is making spinlock acquisition time long enough to trigger a VMWare watchdog. I enabled driver verifier with verifier /rc 1 2 4 5 6 8 9 12 18 20 24 26 35 /driver <mydriver.sys> as mentioned in this NTDEV post. Here's the call stack:
0: kd> kc
# Call Site
00 nt!DbgBreakPointWithStatus
01 nt!KiBugCheckDebugBreak
02 nt!KeBugCheck2
03 nt!KeBugCheckEx
04 nt!HalBugCheckSystem
05 PSHED!PshedBugCheckSystem
06 nt!WheaReportHwError
07 nt!HalHandleNMI
08 nt!KiProcessNMI
09 nt!KxNmiInterrupt
0a nt!KiNmiInterrupt
0b nt!HalpApic1EndOfInterrupt
0c nt!HalPerformEndOfInterrupt
0d nt!KiInterruptDispatchNoLockNoEtw
0e nt!MiWalkVaCheckCommon
0f nt!MiEmptyWorkingSetHelper
10 nt!MiEmptyPte
11 nt!MiWalkPageTablesRecursively
12 nt!MiWalkPageTablesRecursively
13 nt!MiWalkPageTablesRecursively
14 nt!MiWalkPageTablesRecursively
15 nt!MiWalkPageTables
16 nt!MiEmptyWorkingSetInitiate
17 nt!MiEmptyTargetedWorkingSet
18 nt!MiTrimAllSystemPagableMemory
19 nt!MmVerifierTrimMemory
1a nt!ViKeRaiseIrqlSanityChecks
1b nt!ViKeAcquireSpinLockRaiseToDpcCommon
1c nt!VerifierKeAcquireSpinLockRaiseToDpc
My driver is at (omitted) frame 1d. These NMI bugchecks never occur in a release build of my driver AND with driver verifier disabled. Can I just chalk this NMI bugcheck up to long delays due to when driver verifier is enabled?