Conflicting statements in MSDN about IoReleaseRemoveLockAndWait

I’m getting a BSOD ‘sometimes’ in WDF when a backend PnP resource gets removed. I can see the surprise remove and remove coming in my driver, at which point I do call the IoReleaseRemoveLockAndWait() prior to deleting the device object.

I do call IoReleaseRemoveLockAndWait() *before* passing the remove IRP down to the next driver as stated in one MSDN article.

Investigating the BSOD I am wondering if that call should not be made *after* passing the remove IRP down to the next driver and what triggered this thinking of mine was a second MSDN article that says it should be done that way.

So at this point I get conflicting MSDN information and would like to know which sequence is the correct one.

Thank you.
Thierry Laurent

related MSDN articles:

http://msdn.microsoft.com/en-us/library/windows/hardware/ff565504(v=vs.85).aspx

<<
To allow queued I/O operations to complete, each driver should call IoReleaseRemoveLockAndWait before it passes the IRP_MN_REMOVE_DEVICE request to the next-lower driver, and before it releases memory, calls IoDetachDevice, or calls IoDeleteDevice.

>

http://msdn.microsoft.com/en-us/library/windows/hardware/ff549567(v=vs.85).aspx

<<
To allow queued I/O requests to complete, each driver should call IoReleaseRemoveLockAndWait after it passes the remove IRP to the next-lower driver, and before it releases memory, calls IoDetachDevice, or calls IoDeleteDevice.

>

Do you have a WDF driver? If so, why are you using a remove lock? The
framework does a good job of handling this kind of thing for you. Also, any
time you’re posting about a crash you should provide the !analyze -v output.

That’s pretty funny. Note that the second doc item goes on to say:

“The call to IoReleaseRemoveLockAndWait enables the bus driver to cancel any
IRPs that might be outstanding (for example, an IRP_MN_WAIT_WAKE IRP).”

Which is clearly incorrect. Passing the *remove IRP* down allows the bus
driver to cancel outstanding IRPs, but the call to IoReleaseXxx does nothing
of the sort. And people wonder why writing a driver in WDM is so hard…

There are only two remaining WDM samples in the 8.1 WDK and both call
IoReleaseXxx *after* sending the IRP_MN_REMOVE IRP down. OTOH, the Win7 WDK
has a couple of samples that show calling IoReleaseXxx *before* sending the
IRP_MN_REMOVE IRP down.

I suspect that the real answer is that calling IoReleaseXxx *after* sending
the IRP down is always correct and in some cases necessary, as the
documentation alludes to above (e.g. your remove lock might not fire until
the bus driver cancels your wait/wake IRP). Calling it *before* might also
be correct, but that assumes that your remove lock does not rely on any
activity being performed in the bus driver.

Clearly the docs need a bit of cleanup here (filing a bug as we speak).
However, I can’t imagine that choosing to call it before or after has
anything to do with the system crashing, which is why I’d like to see the
!analyze -v output.

-scott
OSR

wrote in message news:xxxxx@ntdev…

I’m getting a BSOD ‘sometimes’ in WDF when a backend PnP resource gets
removed. I can see the surprise remove and remove coming in my driver, at
which point I do call the IoReleaseRemoveLockAndWait() prior to deleting the
device object.

I do call IoReleaseRemoveLockAndWait() *before* passing the remove IRP down
to the next driver as stated in one MSDN article.

Investigating the BSOD I am wondering if that call should not be made
*after* passing the remove IRP down to the next driver and what triggered
this thinking of mine was a second MSDN article that says it should be done
that way.

So at this point I get conflicting MSDN information and would like to know
which sequence is the correct one.

Thank you.
Thierry Laurent

related MSDN articles:

http://msdn.microsoft.com/en-us/library/windows/hardware/ff565504(v=vs.85).aspx

<<
To allow queued I/O operations to complete, each driver should call
IoReleaseRemoveLockAndWait before it passes the IRP_MN_REMOVE_DEVICE request
to the next-lower driver, and before it releases memory, calls
IoDetachDevice, or calls IoDeleteDevice.

>

http://msdn.microsoft.com/en-us/library/windows/hardware/ff549567(v=vs.85).aspx

<<
To allow queued I/O requests to complete, each driver should call
IoReleaseRemoveLockAndWait after it passes the remove IRP to the next-lower
driver, and before it releases memory, calls IoDetachDevice, or calls
IoDeleteDevice.

>

Thank you Scott.

My driver is WDM. I can recreate the BSOD but not regularly. Every time I can though it shows my driver’s stack to be in the call passing the remove IRP down, hence my suspicion on the ordering.
I will put my stack as well but first is the !analyze -v output. I have driver verifier running on the machine and it did not detect anything special.

<<
DRIVER_IRQL_NOT_LESS_OR_EQUAL (d1)
An attempt was made to access a pageable (or completely invalid) address at an
interrupt request level (IRQL) that is too high. This is usually
caused by drivers using improper addresses.
If kernel debugger is available get stack backtrace.
Arguments:
Arg1: ffffffffffffffd0, memory referenced
Arg2: 0000000000000002, IRQL
Arg3: 0000000000000000, value 0 = read operation, 1 = write operation
Arg4: fffff880010d0862, address which referenced memory

Debugging Details:

READ_ADDRESS: ffffffffffffffd0

CURRENT_IRQL: 2

FAULTING_IP:
Wdf01000!FxDevice::Dispatch+22
fffff880`010d0862 4c8b50d0 mov r10,qword ptr [rax-30h]

DEFAULT_BUCKET_ID: VISTA_DRIVER_FAULT

BUGCHECK_STR: 0xD1

PROCESS_NAME: System

TRAP_FRAME: fffff8800203d2c0 – (.trap 0xfffff8800203d2c0)
NOTE: The trap frame does not contain all registers.
Some register values may be zeroed or incorrect.
rax=0000000000000000 rbx=0000000000000000 rcx=fffffa8003d40370
rdx=fffff9800b642e10 rsi=0000000000000000 rdi=0000000000000000
rip=fffff880010d0862 rsp=fffff8800203d450 rbp=0000000000000002
r8=fffffa8002c97e70 r9=0000000000000001 r10=fffff9800b642e10
r11=fffff9800b642e02 r12=0000000000000000 r13=0000000000000000
r14=0000000000000000 r15=0000000000000000
iopl=0 nv up ei pl zr na po nc
Wdf01000!FxDevice::Dispatch+0x22:
fffff880`010d0862 4c8b50d0 mov r10,qword ptr [rax-30h] ds:dc18:ffd0=???
Resetting default scope

LAST_CONTROL_TRANSFER: from fffff80001dc9a12 to fffff80001cd0a30

STACK_TEXT:
fffff8800203ca08 fffff80001dc9a12 : ffffffffffffffd0 fffff8800201ff40 0000000000000065 fffff80001d13888 : nt!RtlpBreakWithStatusInstruction
fffff8800203ca10 fffff80001dca7fe : 0000000000000003 0000000000000000 fffff80001d140e0 00000000000000d1 : nt!KiBugCheckDebugBreak+0x12
fffff8800203ca70 fffff80001cd8d04 : fffffa80036e9e20 fffff88000e60872 0000000000000001 fffff80001d550c4 : nt!KeBugCheck2+0x71e
fffff8800203d140 fffff80001cd81a9 : 000000000000000a ffffffffffffffd0 0000000000000002 0000000000000000 : nt!KeBugCheckEx+0x104
fffff8800203d180 fffff80001cd6e20 : 0000000000000000 00000000a0000003 fffff9801f920e10 fffff9800b642e10 : nt!KiBugCheckDispatch+0x69
fffff8800203d2c0 fffff880010d0862 : fffffa8003d40370 0000000000000002 fffffa800473ff40 fffff8000217ee96 : nt!KiPageFault+0x260
fffff8800203d450 fffff880010d0aa6 : fffff9800b642e10 0000000000000002 fffffa8003d40370 fffff8800203dc18 : Wdf01000!FxDevice::Dispatch+0x22
fffff8800203d490 fffff80002182d26 : fffff9800b642e10 0000000000000002 fffffa8003d40370 e83f03ffffe00007 : Wdf01000!FxDevice::DispatchWithLock+0xa6
fffff8800203d4d0 fffff80001ce384c : fffff8800203d5f0 0000000000000000 0000000000000001 fffffa800473ff40 : nt!IovCallDriver+0x566
fffff8800203d530 fffff80001ce36e6 : fffffa8004360748 fffffa8004360748 0000000000000000 0000000000000000 : nt!KiProcessTimerDpcTable+0x6c
fffff8800203d5a0 fffff80001ce35ce : 0000001479a9100e fffff8800203dc18 0000000000089683 fffff880020185e8 : nt!KiProcessExpiredTimerList+0xc6
fffff8800203dbf0 fffff80001ce33b7 : 0000140b8ff202c5 0000140b00089683 0000140b8ff20263 0000000000000083 : nt!KiTimerExpiration+0x1be
fffff8800203dc90 fffff80001cd090a : fffff88002015180 fffff8800201ff40 0000000000000001 fffff80000000000 : nt!KiRetireDpcList+0x277
fffff8800203dd40 0000000000000000 : fffff8800203e000 fffff88002038000 fffff8800203dd00 0000000000000000 : nt!KiIdleLoop+0x5a

STACK_COMMAND: kb

FOLLOWUP_IP:
Wdf01000!FxDevice::Dispatch+22
fffff880`010d0862 4c8b50d0 mov r10,qword ptr [rax-30h]

SYMBOL_STACK_INDEX: 6

SYMBOL_NAME: Wdf01000!FxDevice::Dispatch+22

FOLLOWUP_NAME: MachineOwner

MODULE_NAME: Wdf01000

IMAGE_NAME: Wdf01000.sys

DEBUG_FLR_IMAGE_TIMESTAMP: 5010aa89

FAILURE_BUCKET_ID: X64_0xD1_VRF_Wdf01000!FxDevice::Dispatch+22

BUCKET_ID: X64_0xD1_VRF_Wdf01000!FxDevice::Dispatch+22

Followup: MachineOwner

>

and the thread’s stack related to the PnP device removal that’s going through my driver (DcsPMF and DcsDf) at the time of the BSOD (note: the drivers send the IRP down and wait for its completion so if STATUS_PENDING is returned there’s a WaitForSingleObject that wakes up when the IRP completion routine sets a private event).

<<
*** Stack trace for last set context - .thread/.cxr resets it
Child-SP RetAddr : Args to Child : Call Site
fffff880059aac50 fffff80001cce612 : 0000000000000017 fffffa80044153c0 0000000000000000 0000000000000008 : nt!KiSwapContext+0x7a
fffff880059aad90 fffff80001cdf98f : fffffa8002ca6001 fffffa80040621b0 0000000000000000 0000000000000000 : nt!KiCommitThreadWait+0x1d2
fffff880059aae20 fffff880013e08fa : 0000000000000000 0000000000000000 fffffa8004062000 fffffa8004062100 : nt!KeWaitForSingleObject+0x19f
fffff880059aaec0 fffff880013dcb08 : fffffa8003d40370 0000000000000000 0000000000000002 fffff880013d7110 : CLASSPNP!ClassRemoveDevice+0x12a
fffff880059aaf00 fffff80002182d26 : fffffa80040621b0 fffffa8002ca6038 fffff9801f614c60 fffff8000217e48e : CLASSPNP! ?? ::NNGAKEGL::string'+0x17a6 fffff880059aaf90 fffff800021855ea : fffff9801f614e50 fffffa8004062060 fffffa80044372a0 fffffa80018fc320 : nt!IovCallDriver+0x566 fffff880059aaff0 fffff80002182d26 : fffff9801f614c60 0000000000000002 fffffa80044372a0 fffffa800421a0b0 : nt!ViFilterDispatchPnp+0xea fffff880059ab020 fffff880017e270c : fffff9801f614c60 0000000000000002 fffff9801f614f18 fffffa800421a0b0 : nt!IovCallDriver+0x566 fffff880059ab080 fffff880017e6298 : fffffa8002d26830 fffff9801f614c60 fffffa8002d26830 fffffa80032a1410 : DcsDf!DiskPerfForwardIrpSynchronous+0xac fffff880059ab110 fffff880017e5778 : fffffa8002d26830 fffff9801f614c60 0000000000000000 0000000000000000 : DcsDf!DiskPerfRemoveDevice+0x58 fffff880059ab150 fffff80002182d26 : fffffa8002d26830 fffff9801f614c60 fffff9801f614c60 fffff8000217e48e : DcsDf!DiskPerfDispatchPnp+0x148 fffff880059ab1d0 fffff800021855ea : fffff9801f614ee0 fffffa8002d26830 fffffa80042f5af0 fffffa80032a1410 : nt!IovCallDriver+0x566 fffff880059ab230 fffff80002182d26 : fffff9801f614c60 0000000000000002 fffffa80042f5af0 fffffa8003212a70 : nt!ViFilterDispatchPnp+0xea fffff880059ab260 fffff880014b6926 : fffff9801f614c60 0000000000000002 fffff9801f614fa8 fffffa8003212a70 : nt!IovCallDriver+0x566 fffff880059ab2c0 fffff880014b82c5 : fffffa8003405060 fffff9801f614c60 fffff880059ab3f0 fffff880059ab3d8 : DcsPMF!PMFForwardIrpSynchronous+0xd6 fffff880059ab340 fffff880014b8d78 : fffffa8003405060 fffff9801f614c60 0000000000000003 0000000000000000 : DcsPMF!PMFRemoveDevice+0x2e5 fffff880059ab510 fffff80002182d26 : fffffa8003405060 fffff9801f614c60 fffff9801f614c60 fffff8000217e48e : DcsPMF!PMFDispatchPnp+0x738 fffff880059ab5d0 fffff800021855ea : fffff9801f614f70 fffffa8003405060 fffffa8003fd2700 fffffa80043d6eb0 : nt!IovCallDriver+0x566 fffff880059ab630 fffff80002182d26 : fffff9801f614c60 0000000000000002 fffffa8003fd2700 fffffa8003b2a010 : nt!ViFilterDispatchPnp+0xea fffff880059ab660 fffff8800108530e : fffff9801f614c60 fffff9801f614fb8 0000000000000000 fffffa8003b2a010 : nt!IovCallDriver+0x566 fffff880059ab6c0 fffff80002182d26 : fffff9801f614c60 0000000000000002 fffffa80039c0330 00000000c00000bb : partmgr!PmPnp+0xfe fffff880059ab710 fffff80001f46d11 : fffffa80039c0330 fffff880059ab828 00000000c00000bb fffffa8003b44010 : nt!IovCallDriver+0x566 fffff880059ab770 fffff800020c5681 : fffffa8004743060 0000000000000000 fffffa80047384a0 0000000000000801 : nt!IopSynchronousCall+0xe1 fffff880059ab7e0 fffff80001ddb063 : fffff8a0023b6f20 fffff8a0023b6f20 0000000000000018 0000000000000000 : nt!IopRemoveDevice+0x101 fffff880059ab8a0 fffff800020c51d4 : fffffa80047384a0 0000000000000000 0000000000000002 0000000000000000 : nt!PnpRemoveLockedDeviceNode+0x1a3 fffff880059ab8f0 fffff800020c52e0 : 0000000000000000 fffffa8004743000 fffff8a0074b8900 fffff80001edca80 : nt!PnpDeleteLockedDeviceNode+0x44 fffff880059ab920 fffff800020c53d9 : fffffa8004601602 fffffa8004601600 0000000000000001 0000000000000000 : nt!PnpDeleteLockedDeviceNodes+0xa0 fffff880059ab990 fffff800020c5551 : fffffa8004601600 0000000000000000 fffffa8004601600 0000000000010286 : nt!PnpDelayedRemoveWorker+0x79 fffff880059ab9e0 fffff80001ddb289 : 0000000000000000 fffffa80035e0a50 0000000000000001 0000000000000000 : nt!PnpChainDereferenceComplete+0x131 fffff880059aba20 fffff80002156570 : 0000000000000000 fffffa8004738400 fffff8a00398cef0 0000000000000001 : nt!PnpIsChainDereferenced+0xc9 fffff880059abaa0 fffff8000215680c : fffff880059abc78 0000000000010200 fffff880059abc00 0000000000000000 : nt!PnpProcessQueryRemoveAndEject+0xff0 fffff880059abbe0 fffff8000203f9ae : 0000000000000000 fffffa800466eb00 fffff8a00398cef0 fffff880059abc00 : nt!PnpProcessTargetDeviceEvent+0x4c fffff880059abc10 fffff80001ce2251 : fffff80001f45b98 fffff8a00398cef0 fffff80001e7e2d8 fffff80001e7e2d8 : nt! ?? ::NNGAKEGL::string’+0x552ab
fffff880059abc70 fffff80001f76ede : 0000000000000000 fffffa80044153c0 0000000000000080 fffffa8001876890 : nt!ExpWorkerThread+0x111
fffff880059abd00 fffff80001cc9906 : fffff88002015180 fffffa80044153c0 fffffa80028dcb50 0000000000000202 : nt!PspSystemThreadStartup+0x5a
fffff880059abd40 0000000000000000 : fffff880059ac000 fffff880059a6000 fffff880059aa8a0 0000000000000000 : nt!KxStartSystemThread+0x16

>

Do you have any timer DPCs in your drivers? It looks to me like someone is
calling into a WDF driver after the device has been removed. From the stack
I can’t tell what either of the drivers are though (tail call optimization
obscures the caller, the framework owning the entry points obscures the
callee).

I set a breakpoint on this routine on my test system and at this point RCX
should contain a device object (RCX+40 is the DeviceExtension field, WDF is
trying to access some header before this field). !devobj @rcx or "dt
nt!_device_object @rcx’ might tell you the driver, which would shed some
light on things.

-scott
OSR

wrote in message news:xxxxx@ntdev…

Thank you Scott.

My driver is WDM. I can recreate the BSOD but not regularly. Every time I
can though it shows my driver’s stack to be in the call passing the remove
IRP down, hence my suspicion on the ordering.
I will put my stack as well but first is the !analyze -v output. I have
driver verifier running on the machine and it did not detect anything
special.

<<
DRIVER_IRQL_NOT_LESS_OR_EQUAL (d1)
An attempt was made to access a pageable (or completely invalid) address at
an
interrupt request level (IRQL) that is too high. This is usually
caused by drivers using improper addresses.
If kernel debugger is available get stack backtrace.
Arguments:
Arg1: ffffffffffffffd0, memory referenced
Arg2: 0000000000000002, IRQL
Arg3: 0000000000000000, value 0 = read operation, 1 = write operation
Arg4: fffff880010d0862, address which referenced memory

Debugging Details:

READ_ADDRESS: ffffffffffffffd0

CURRENT_IRQL: 2

FAULTING_IP:
Wdf01000!FxDevice::Dispatch+22
fffff880`010d0862 4c8b50d0 mov r10,qword ptr [rax-30h]

DEFAULT_BUCKET_ID: VISTA_DRIVER_FAULT

BUGCHECK_STR: 0xD1

PROCESS_NAME: System

TRAP_FRAME: fffff8800203d2c0 – (.trap 0xfffff8800203d2c0)
NOTE: The trap frame does not contain all registers.
Some register values may be zeroed or incorrect.
rax=0000000000000000 rbx=0000000000000000 rcx=fffffa8003d40370
rdx=fffff9800b642e10 rsi=0000000000000000 rdi=0000000000000000
rip=fffff880010d0862 rsp=fffff8800203d450 rbp=0000000000000002
r8=fffffa8002c97e70 r9=0000000000000001 r10=fffff9800b642e10
r11=fffff9800b642e02 r12=0000000000000000 r13=0000000000000000
r14=0000000000000000 r15=0000000000000000
iopl=0 nv up ei pl zr na po nc
Wdf01000!FxDevice::Dispatch+0x22:
fffff880`010d0862 4c8b50d0 mov r10,qword ptr [rax-30h]
ds:dc18:ffd0=???
Resetting default scope

LAST_CONTROL_TRANSFER: from fffff80001dc9a12 to fffff80001cd0a30

STACK_TEXT:
fffff8800203ca08 fffff80001dc9a12 : ffffffffffffffd0 fffff8800201ff40
0000000000000065 fffff80001d13888 : nt!RtlpBreakWithStatusInstruction
fffff8800203ca10 fffff80001dca7fe : 0000000000000003 0000000000000000
fffff80001d140e0 00000000000000d1 : nt!KiBugCheckDebugBreak+0x12
fffff8800203ca70 fffff80001cd8d04 : fffffa80036e9e20 fffff88000e60872
0000000000000001 fffff80001d550c4 : nt!KeBugCheck2+0x71e
fffff8800203d140 fffff80001cd81a9 : 000000000000000a ffffffffffffffd0
0000000000000002 0000000000000000 : nt!KeBugCheckEx+0x104
fffff8800203d180 fffff80001cd6e20 : 0000000000000000 00000000a0000003
fffff9801f920e10 fffff9800b642e10 : nt!KiBugCheckDispatch+0x69
fffff8800203d2c0 fffff880010d0862 : fffffa8003d40370 0000000000000002
fffffa800473ff40 fffff8000217ee96 : nt!KiPageFault+0x260
fffff8800203d450 fffff880010d0aa6 : fffff9800b642e10 0000000000000002
fffffa8003d40370 fffff8800203dc18 : Wdf01000!FxDevice::Dispatch+0x22
fffff8800203d490 fffff80002182d26 : fffff9800b642e10 0000000000000002
fffffa8003d40370 e83f03ffffe00007 :
Wdf01000!FxDevice::DispatchWithLock+0xa6
fffff8800203d4d0 fffff80001ce384c : fffff8800203d5f0 0000000000000000
0000000000000001 fffffa800473ff40 : nt!IovCallDriver+0x566
fffff8800203d530 fffff80001ce36e6 : fffffa8004360748 fffffa8004360748
0000000000000000 0000000000000000 : nt!KiProcessTimerDpcTable+0x6c
fffff8800203d5a0 fffff80001ce35ce : 0000001479a9100e fffff8800203dc18
0000000000089683 fffff880020185e8 : nt!KiProcessExpiredTimerList+0xc6
fffff8800203dbf0 fffff80001ce33b7 : 0000140b8ff202c5 0000140b00089683
0000140b8ff20263 0000000000000083 : nt!KiTimerExpiration+0x1be
fffff8800203dc90 fffff80001cd090a : fffff88002015180 fffff8800201ff40
0000000000000001 fffff80000000000 : nt!KiRetireDpcList+0x277
fffff8800203dd40 0000000000000000 : fffff8800203e000 fffff88002038000
fffff8800203dd00 0000000000000000 : nt!KiIdleLoop+0x5a

STACK_COMMAND: kb

FOLLOWUP_IP:
Wdf01000!FxDevice::Dispatch+22
fffff880`010d0862 4c8b50d0 mov r10,qword ptr [rax-30h]

SYMBOL_STACK_INDEX: 6

SYMBOL_NAME: Wdf01000!FxDevice::Dispatch+22

FOLLOWUP_NAME: MachineOwner

MODULE_NAME: Wdf01000

IMAGE_NAME: Wdf01000.sys

DEBUG_FLR_IMAGE_TIMESTAMP: 5010aa89

FAILURE_BUCKET_ID: X64_0xD1_VRF_Wdf01000!FxDevice::Dispatch+22

BUCKET_ID: X64_0xD1_VRF_Wdf01000!FxDevice::Dispatch+22

Followup: MachineOwner

>

and the thread’s stack related to the PnP device removal that’s going
through my driver (DcsPMF and DcsDf) at the time of the BSOD (note: the
drivers send the IRP down and wait for its completion so if STATUS_PENDING
is returned there’s a WaitForSingleObject that wakes up when the IRP
completion routine sets a private event).

<<
*** Stack trace for last set context - .thread/.cxr resets it
Child-SP RetAddr : Args to Child
: Call Site
fffff880059aac50 fffff80001cce612 : 0000000000000017 fffffa80044153c0
0000000000000000 0000000000000008 : nt!KiSwapContext+0x7a
fffff880059aad90 fffff80001cdf98f : fffffa8002ca6001 fffffa80040621b0
0000000000000000 0000000000000000 : nt!KiCommitThreadWait+0x1d2
fffff880059aae20 fffff880013e08fa : 0000000000000000 0000000000000000
fffffa8004062000 fffffa8004062100 : nt!KeWaitForSingleObject+0x19f
fffff880059aaec0 fffff880013dcb08 : fffffa8003d40370 0000000000000000
0000000000000002 fffff880013d7110 : CLASSPNP!ClassRemoveDevice+0x12a
fffff880059aaf00 fffff80002182d26 : fffffa80040621b0 fffffa8002ca6038
fffff9801f614c60 fffff8000217e48e : CLASSPNP! ??
::NNGAKEGL::string'+0x17a6 fffff880059aaf90 fffff800021855ea : fffff9801f614e50 fffffa8004062060 fffffa80044372a0 fffffa80018fc320 : nt!IovCallDriver+0x566 fffff880059aaff0 fffff80002182d26 : fffff9801f614c60 0000000000000002 fffffa80044372a0 fffffa800421a0b0 : nt!ViFilterDispatchPnp+0xea fffff880059ab020 fffff880017e270c : fffff9801f614c60 0000000000000002 fffff9801f614f18 fffffa800421a0b0 : nt!IovCallDriver+0x566 fffff880059ab080 fffff880017e6298 : fffffa8002d26830 fffff9801f614c60 fffffa8002d26830 fffffa80032a1410 : DcsDf!DiskPerfForwardIrpSynchronous+0xac fffff880059ab110 fffff880017e5778 : fffffa8002d26830 fffff9801f614c60 0000000000000000 0000000000000000 : DcsDf!DiskPerfRemoveDevice+0x58 fffff880059ab150 fffff80002182d26 : fffffa8002d26830 fffff9801f614c60 fffff9801f614c60 fffff8000217e48e : DcsDf!DiskPerfDispatchPnp+0x148 fffff880059ab1d0 fffff800021855ea : fffff9801f614ee0 fffffa8002d26830 fffffa80042f5af0 fffffa80032a1410 : nt!IovCallDriver+0x566 fffff880059ab230 fffff80002182d26 : fffff9801f614c60 0000000000000002 fffffa80042f5af0 fffffa8003212a70 : nt!ViFilterDispatchPnp+0xea fffff880059ab260 fffff880014b6926 : fffff9801f614c60 0000000000000002 fffff9801f614fa8 fffffa8003212a70 : nt!IovCallDriver+0x566 fffff880059ab2c0 fffff880014b82c5 : fffffa8003405060 fffff9801f614c60 fffff880059ab3f0 fffff880059ab3d8 : DcsPMF!PMFForwardIrpSynchronous+0xd6 fffff880059ab340 fffff880014b8d78 : fffffa8003405060 fffff9801f614c60 0000000000000003 0000000000000000 : DcsPMF!PMFRemoveDevice+0x2e5 fffff880059ab510 fffff80002182d26 : fffffa8003405060 fffff9801f614c60 fffff9801f614c60 fffff8000217e48e : DcsPMF!PMFDispatchPnp+0x738 fffff880059ab5d0 fffff800021855ea : fffff9801f614f70 fffffa8003405060 fffffa8003fd2700 fffffa80043d6eb0 : nt!IovCallDriver+0x566 fffff880059ab630 fffff80002182d26 : fffff9801f614c60 0000000000000002 fffffa8003fd2700 fffffa8003b2a010 : nt!ViFilterDispatchPnp+0xea fffff880059ab660 fffff8800108530e : fffff9801f614c60 fffff9801f614fb8 0000000000000000 fffffa8003b2a010 : nt!IovCallDriver+0x566 fffff880059ab6c0 fffff80002182d26 : fffff9801f614c60 0000000000000002 fffffa80039c0330 00000000c00000bb : partmgr!PmPnp+0xfe fffff880059ab710 fffff80001f46d11 : fffffa80039c0330 fffff880059ab828 00000000c00000bb fffffa8003b44010 : nt!IovCallDriver+0x566 fffff880059ab770 fffff800020c5681 : fffffa8004743060 0000000000000000 fffffa80047384a0 0000000000000801 : nt!IopSynchronousCall+0xe1 fffff880059ab7e0 fffff80001ddb063 : fffff8a0023b6f20 fffff8a0023b6f20 0000000000000018 0000000000000000 : nt!IopRemoveDevice+0x101 fffff880059ab8a0 fffff800020c51d4 : fffffa80047384a0 0000000000000000 0000000000000002 0000000000000000 : nt!PnpRemoveLockedDeviceNode+0x1a3 fffff880059ab8f0 fffff800020c52e0 : 0000000000000000 fffffa8004743000 fffff8a0074b8900 fffff80001edca80 : nt!PnpDeleteLockedDeviceNode+0x44 fffff880059ab920 fffff800020c53d9 : fffffa8004601602 fffffa8004601600 0000000000000001 0000000000000000 : nt!PnpDeleteLockedDeviceNodes+0xa0 fffff880059ab990 fffff800020c5551 : fffffa8004601600 0000000000000000 fffffa8004601600 0000000000010286 : nt!PnpDelayedRemoveWorker+0x79 fffff880059ab9e0 fffff80001ddb289 : 0000000000000000 fffffa80035e0a50 0000000000000001 0000000000000000 : nt!PnpChainDereferenceComplete+0x131 fffff880059aba20 fffff80002156570 : 0000000000000000 fffffa8004738400 fffff8a00398cef0 0000000000000001 : nt!PnpIsChainDereferenced+0xc9 fffff880059abaa0 fffff8000215680c : fffff880059abc78 0000000000010200 fffff880059abc00 0000000000000000 : nt!PnpProcessQueryRemoveAndEject+0xff0 fffff880059abbe0 fffff8000203f9ae : 0000000000000000 fffffa800466eb00 fffff8a00398cef0 fffff880059abc00 : nt!PnpProcessTargetDeviceEvent+0x4c fffff880059abc10 fffff80001ce2251 : fffff80001f45b98 fffff8a00398cef0 fffff80001e7e2d8 fffff80001e7e2d8 : nt! ?? ::NNGAKEGL::string’+0x552ab
fffff880059abc70 fffff80001f76ede : 0000000000000000 fffffa80044153c0
0000000000000080 fffffa8001876890 : nt!ExpWorkerThread+0x111
fffff880059abd00 fffff80001cc9906 : fffff88002015180 fffffa80044153c0
fffffa80028dcb50 0000000000000202 : nt!PspSystemThreadStartup+0x5a
fffff880059abd40 0000000000000000 : fffff880059ac000 fffff880059a6000
fffff880059aa8a0 0000000000000000 : nt!KxStartSystemThread+0x16

>