Windows 10睡眠/休眠后触发DRIVER_POWER_STATE_FAILURE崩溃，故障设备转移至GPU

我遇到了一个奇怪的问题：电脑进入睡眠或休眠状态后，显示器会关闭，过一段时间就会崩溃。而且这个问题不是每次都出现——开机后的头几个小时里，睡眠/休眠功能完全正常，只有开机一段时间后才会触发崩溃。这是硬件层面的崩溃，系统会生成转储文件，而且在我尝试解决的过程中，故障设备还发生了转移，下面是两次崩溃的转储分析供参考。

第一次崩溃转储分析 #

*******************************************************************************

*                                                                             *

*                        Bugcheck Analysis                                    *

*                                                                             *

*******************************************************************************

DRIVER_POWER_STATE_FAILURE (9f)

A driver has failed to complete a power IRP within a specific time.

Arguments:

Arg1: 0000000000000003, A device object has been blocking an Irp for too long a time

Arg2: ffffd50f008a22f0, Physical Device Object of the stack

Arg3: fffff8076c6f4850, nt!TRIAGE_9F_POWER on Win7 and higher, otherwise the Functional Device Object of the stack

Arg4: ffffd50f076f7de0, The blocked IRP

Debugging Details:

------------------

KEY_VALUES_STRING: 1

Key  : Analysis.CPU.Sec

Value: 2

Key  : Analysis.DebugAnalysisProvider.CPP

Value: Create: 8007007e on CYGNI

Key  : Analysis.DebugData

Value: CreateObject

Key  : Analysis.DebugModel

Value: CreateObject

Key  : Analysis.Elapsed.Sec

Value: 1

Key  : Analysis.Memory.CommitPeak.Mb

Value: 80

Key  : Analysis.System

Value: CreateObject

Key  : Dump.Attributes.InsufficientDumpfileSize

Value: 1

DUMP_FILE_ATTRIBUTES: 0xc

Insufficient Dumpfile Size

Kernel Generated Triage Dump

BUGCHECK_CODE:  9f

BUGCHECK_P1: 3

BUGCHECK_P2: ffffd50f008a22f0

BUGCHECK_P3: fffff8076c6f4850

BUGCHECK_P4: ffffd50f076f7de0

DRVPOWERSTATE_SUBCODE:  3

IMAGE_NAME:  pci.sys

MODULE_NAME: pci

FAULTING_MODULE: fffff8076b260000 pci

BLACKBOXBSD: 1 (!blackboxbsd)

BLACKBOXNTFS: 1 (!blackboxntfs)

BLACKBOXPNP: 1 (!blackboxpnp)

BLACKBOXWINLOGON: 1

CUSTOMER_CRASH_COUNT:  1

PROCESS_NAME:  System

STACK_TEXT:

fffff807`6c6f4818 fffff807`665637a7 : 00000000`0000009f 00000000`00000003 ffffd50f`008a22f0 fffff807`6c6f4850 : nt!KeBugCheckEx

fffff807`6c6f4820 fffff807`665636c1 : ffffd50f`246a6d70 00000000`00000000 ffffd50f`1b4a41f0 00000000`00000000 : nt!PopIrpWatchdogBugcheck+0xdf

fffff807`6c6f4890 fffff807`6628bbd2 : ffffd50f`246a6da8 fffff807`641c9180 fffff807`641c9180 00000000`00000002 : nt!PopIrpWatchdog+0x31

fffff807`6c6f48e0 fffff807`6628ac4d : 00000000`00000000 fffff807`662d9485 00000000`00140001 00000000`000a5e26 : nt!KiProcessExpiredTimerList+0x172

fffff807`6c6f49d0 fffff807`6640059e : 00000000`00000000 fffff807`641c9180 fffff807`66d27a00 ffffd50f`2112e440 : nt!KiRetireDpcList+0x5dd

fffff807`6c6f4c60 00000000`00000000 : fffff807`6c6f5000 fffff807`6c6ef000 00000000`00000000 00000000`00000000 : nt!KiIdleLoop+0x9e

IMAGE_VERSION:  10.0.19041.2728

STACK_COMMAND:  .thread ; .cxr ; kb

FAILURE_BUCKET_ID:  0x9F_3_POWER_DOWN_rt640x64_IMAGE_pci.sys

OSPLATFORM_TYPE:  x64

OSNAME:  Windows 10

FAILURE_ID_HASH:  {2ce71122-4407-ac30-726e-512dc09f0103}

Followup:     MachineOwner

运行!devstack后的设备栈信息 #

0: kd> !devstack ffffd50f008a22f0

!DevObj           !DrvObj            !DevExt           ObjectName

ffffd50f050c8050 Unable to load image rt640x64.sys, Win32 error 0n2

*** WARNING: Unable to verify timestamp for rt640x64.sys

\Driver\rt640x64   ffffd50f050c81a0  InfoMask field not found for _OBJECT_HEADER at ffffd50f050c8020

> ffffd50f008a22f0  \Driver\pci        ffffd50f008a2440  Cannot read info offset from nt!ObpInfoMaskToOffset

!DevNode ffffd50f008a2c30 :

DeviceInst is "PCI\VEN_10EC&DEV_8168&SUBSYS_E0001458&REV_0C\01000000684CE00000"

ServiceName is "rt640x64"

根据设备ID和厂商ID，这是一块Realtek PCIe千兆以太网适配器。我禁用了这个设备，但问题并没有解决，反而故障转移到了GPU上：

第二次崩溃转储分析 #

*******************************************************************************

*                                                                             *

*                        Bugcheck Analysis                                    *

*                                                                             *

*******************************************************************************

DRIVER_POWER_STATE_FAILURE (9f)

A driver has failed to complete a power IRP within a specific time.

Arguments:

Arg1: 0000000000000003, A device object has been blocking an Irp for too long a time

Arg2: ffffd68fa3676120, Physical Device Object of the stack

Arg3: fffff806820f4850, nt!TRIAGE_9F_POWER on Win7 and higher, otherwise the Functional Device Object of the stack

Arg4: ffffd68fadb688d0, The blocked IRP

Debugging Details:

------------------

KEY_VALUES_STRING: 1

Key  : Analysis.CPU.Sec

Value: 2

Key  : Analysis.DebugAnalysisProvider.CPP

Value: Create: 8007007e on Desktop

Key  : Analysis.DebugData

Value: CreateObject

Key  : Analysis.DebugModel

Value: CreateObject

Key  : Analysis.Elapsed.Sec

Value: 3

Key  : Analysis.Memory.CommitPeak.Mb

Value: 85

Key  : Analysis.System

Value: CreateObject

Key  : Dump.Attributes.InsufficientDumpfileSize

Value: 1

DUMP_FILE_ATTRIBUTES: 0xc

Insufficient Dumpfile Size

Kernel Generated Triage Dump

BUGCHECK_CODE:  9f

BUGCHECK_P1: 3

BUGCHECK_P2: ffffd68fa3676120

BUGCHECK_P3: fffff806820f4850

BUGCHECK_P4: ffffd68fadb688d0

DRVPOWERSTATE_SUBCODE:  3

IMAGE_NAME:  pci.sys

MODULE_NAME: pci

FAULTING_MODULE: fffff80680e60000 pci

BLACKBOXBSD: 1 (!blackboxbsd)

BLACKBOXNTFS: 1 (!blackboxntfs)

BLACKBOXPNP: 1 (!blackboxpnp)

BLACKBOXWINLOGON: 1

CUSTOMER_CRASH_COUNT:  1

PROCESS_NAME:  System

STACK_TEXT:

fffff806`820f4818 fffff806`7c7637a7 : 00000000`0000009f 00000000`00000003 ffffd68f`a3676120 fffff806`820f4850 : nt!KeBugCheckEx

fffff806`820f4820 fffff806`7c7636c1 : ffffd68f`b5ab72d0 ffffd68f`c4781080 00000000`00000000 00000000`0022d693 : nt!PopIrpWatchdogBugcheck+0xdf

fffff806`820f4890 fffff806`7c48bbd2 : ffffd68f`b5ab7308 fffff806`79cff180 fffff806`820f4b18 00000000`00000000 : nt!PopIrpWatchdog+0x31

fffff806`820f48e0 fffff806`7c48ac4d : 00000000`00000000 fffff806`7c4d9485 00000000`00140001 00000000`0022d693 : nt!KiProcessExpiredTimerList+0x172

fffff806`820f49d0 fffff806`7c60059e : 00000000`00000000 fffff806`79cff180 fffff806`7cf27a00 ffffd68f`ad1a5040 : nt!KiRetireDpcList+0x5dd

fffff806`820f4c60 00000000`00000000 : fffff806`820f5000 fffff806`820ef000 00000000`00000000 00000000`00000000 : nt!KiIdleLoop+0x9e

IMAGE_VERSION:  10.0.19041.2728

STACK_COMMAND:  .thread ; .cxr ; kb

FAILURE_BUCKET_ID:  0x9F_3_POWER_DOWN_nvlddmkm_IMAGE_pci.sys

OSPLATFORM_TYPE:  x64

OSNAME:  Windows 10

FAILURE_ID_HASH:  {5337281a-2de3-9ebb-5277-3f179de9b48f}

Followup:     MachineOwner

运行!devstack后的设备栈信息 #

0: kd> !devstack ffffd68fa3676120

!DevObj           !DrvObj            !DevExt           ObjectName

ffffd68fa7164030 Unable to load image nvlddmkm.sys, Win32 error 0n2

*** WARNING: Unable to verify timestamp for nvlddmkm.sys

\Driver\nvlddmkm   ffffd68fa7164180  InfoMask field not found for _OBJECT_HEADER at ffffd68fa7164000

> ffffd68fa3676120  \Driver\pci        ffffd68fa3676270  Cannot read info offset from nt!ObpInfoMaskToOffset

!DevNode ffffd68fa3676a60 :

DeviceInst is "PCI\VEN_10DE&DEV_2187&SUBSYS_87671043&REV_A1\4&1c3d25bb&0&0019"

ServiceName is "nvlddmkm"

已尝试的解决方法 #

我已经试过以下几种方案，但都没有解决问题：

• 将Windows更新到22H2版本
• 运行DISM.exe和sfc /scannow修复系统文件
• 使用ISO镜像执行就地升级
• 用DDU工具彻底卸载GPU驱动后重新安装
• 将“链接状态电源管理”设置为“中等节能”
• 确保设备管理器中所有设备都允许Windows关闭以节省电源

备注：内容来源于stack exchange，提问作者explodingfilms101