BSOD related to NVIDIA Graphics Driver

Hello,

I have been using my laptop normally, until today, when suddenly I got a BSOD PAGE_FAULT_IN_NONPAGED_AREA. I checked the minidump file with BlueScreenViewer, and it only pointed out "ntoskrnl.exe" as the cause. I don't believe that was the case, as I remembered for sure when I looked at the BSOD, there was a line "What caused: " that tells me it was NVIDIA's driver (nvlddmkm.sys) that was the cause for BSOD to happen.

I have tried stress testing my GPU with Furmark for 10 minutes, with the resolution sets for my native monitor resolution for no issues at all. During the test, I also let some games running in the background, and even those ran perfectly without problems.

My Windows Version number is 22H2, OS build 19045.3155. In addition, I include the compressed file below. I look forward to hear the solutions soon, thank you!

It's certainly nvlddmkm.sys that caused this BSOD, you can see that in the trap frame...

Code:

TRAP_FRAME:  ffffab0b4c947160 -- (.trap 0xffffab0b4c947160)
NOTE: The trap frame does not contain all registers.
Some register values may be zeroed or incorrect.
rax=0000000000000001 rbx=0000000000000000 rcx=0000000000000000
rdx=0000000000000000 rsi=0000000000000000 rdi=0000000000000000
rip=fffff8066789681d rsp=ffffab0b4c9472f0 rbp=0000000000000001
 r8=0000000000000000  r9=0000000000000000 r10=0000000000000001
r11=ffffa27e89c00000 r12=0000000000000000 r13=0000000000000000
r14=0000000000000000 r15=0000000000000000
iopl=0         nv up ei pl nz na po nc
nvlddmkm+0xfb681d:
fffff806`6789681d 498b14d3        mov     rdx,qword ptr [r11+rdx*8] ds:ffffa27e`89c00000=????????????????

The mov instruction at nvlddmkm+0xfb681d has referenced an invalid memory location (indicated by the ????????????????). It's entirely possible (even likely) that this is an error in the nvlddmkm.sys driver, but there are some other clues in the dump that suggest that this could be a graphics card hardware issue.

In the full call stack (which I've truncated here for space) we can see the following function calls...

Code:

ffffab0b`4c947748  fffff806`7fbbb943 dxgmms2!VidSchiResetHwEngine+0x1a3
ffffab0b`4c9477b8  fffff806`58b145f0 dxgkrnl!TdrAllowToDebugEngineTimeout
ffffab0b`4c947830  fffff806`58b145f0 dxgkrnl!TdrAllowToDebugEngineTimeout
ffffab0b`4c947898  fffff806`58c63734 watchdog!WdIsDebuggerPresent+0x14
ffffab0b`4c9478f8  fffff806`7fc511be dxgmms2!VidSchiResetEngines+0xaa

Note that you read this from the bottom up. The dxgmms2 functions are part of the Windows DirectX driver, notice the dxgmms2!VidSchiResetEngines function call. That's because there was a problem communicating with the graphics card and a driver reset has been done. The dxgkrnl functions are also part of the Windows DirectX driver, notice the two TDR function calls. TDR is the graphics Timeout Detection and Recovery feature, it detects graphics hangs and resets the driver and adapter to prevent a hang. At the top here we see the dxgmms2 function resetting the graphics card (dxgmms2!VidSchiResetHwEngine).

It's clear from this that the TDR function was triggered during this graphics operation, that could be because of a driver error (nvlddmkm.sys) but it may also be a graphics card problem.

The version of nvlddmkm.sys that you have installed is recent, dating from 24th June 2023...

Code:

4: kd> lmDvmnvlddmkm
Browse full module list
start             end                 module name
fffff806`668e0000 fffff806`6a233000   nvlddmkm T (no symbols)           
    Loaded symbol image file: nvlddmkm.sys
    Image path: \SystemRoot\System32\DriverStore\FileRepository\nvlti.inf_amd64_f313b6b92f428254\nvlddmkm.sys
    Image name: nvlddmkm.sys
    Browse all global symbols  functions  data
    Timestamp:        Sat Jun 24 00:22:34 2023 (64960D1A)
    CheckSum:         0384DE96
    ImageSize:        03953000
    Translations:     0000.04b0 0000.04e4 0409.04b0 0409.04e4
    Information from resource tables:

The best way forward is to download the latest driver for your card, and the two immediately prior versions. Also download the Display Driver Uninstaller DDU. Install each of those three driver versions in turn, using DDU each time to completely remove the previous driver version. If you still see this BSDO with all three driver versions then it's almost certainly a hardware issue with the graphics card .

ubuysa said:

It's certainly nvlddmkm.sys that caused this BSOD, you can see that in the trap frame...

Code:

TRAP_FRAME:  ffffab0b4c947160 -- (.trap 0xffffab0b4c947160)
NOTE: The trap frame does not contain all registers.
Some register values may be zeroed or incorrect.
rax=0000000000000001 rbx=0000000000000000 rcx=0000000000000000
rdx=0000000000000000 rsi=0000000000000000 rdi=0000000000000000
rip=fffff8066789681d rsp=ffffab0b4c9472f0 rbp=0000000000000001
 r8=0000000000000000  r9=0000000000000000 r10=0000000000000001
r11=ffffa27e89c00000 r12=0000000000000000 r13=0000000000000000
r14=0000000000000000 r15=0000000000000000
iopl=0         nv up ei pl nz na po nc
nvlddmkm+0xfb681d:
fffff806`6789681d 498b14d3        mov     rdx,qword ptr [r11+rdx*8] ds:ffffa27e`89c00000=????????????????

The mov instruction at nvlddmkm+0xfb681d has referenced an invalid memory location (indicated by the ????????????????). It's entirely possible (even likely) that this is an error in the nvlddmkm.sys driver, but there are some other clues in the dump that suggest that this could be a graphics card hardware issue.

In the full call stack (which I've truncated here for space) we can see the following function calls...

Code:

ffffab0b`4c947748  fffff806`7fbbb943 dxgmms2!VidSchiResetHwEngine+0x1a3
ffffab0b`4c9477b8  fffff806`58b145f0 dxgkrnl!TdrAllowToDebugEngineTimeout
ffffab0b`4c947830  fffff806`58b145f0 dxgkrnl!TdrAllowToDebugEngineTimeout
ffffab0b`4c947898  fffff806`58c63734 watchdog!WdIsDebuggerPresent+0x14
ffffab0b`4c9478f8  fffff806`7fc511be dxgmms2!VidSchiResetEngines+0xaa

Note that you read this from the bottom up. The dxgmms2 functions are part of the Windows DirectX driver, notice the dxgmms2!VidSchiResetEngines function call. That's because there was a problem communicating with the graphics card and a driver reset has been done. The dxgkrnl functions are also part of the Windows DirectX driver, notice the two TDR function calls. TDR is the graphics Timeout Detection and Recovery feature, it detects graphics hangs and resets the driver and adapter to prevent a hang. At the top here we see the dxgmms2 function resetting the graphics card (dxgmms2!VidSchiResetHwEngine).

It's clear from this that the TDR function was triggered during this graphics operation, that could be because of a driver error (nvlddmkm.sys) but it may also be a graphics card problem.

The version of nvlddmkm.sys that you have installed is recent, dating from 24th June 2023...

Code:

4: kd> lmDvmnvlddmkm
Browse full module list
start             end                 module name
fffff806`668e0000 fffff806`6a233000   nvlddmkm T (no symbols)           
    Loaded symbol image file: nvlddmkm.sys
    Image path: \SystemRoot\System32\DriverStore\FileRepository\nvlti.inf_amd64_f313b6b92f428254\nvlddmkm.sys
    Image name: nvlddmkm.sys
    Browse all global symbols  functions  data
    Timestamp:        Sat Jun 24 00:22:34 2023 (64960D1A)
    CheckSum:         0384DE96
    ImageSize:        03953000
    Translations:     0000.04b0 0000.04e4 0409.04b0 0409.04e4
    Information from resource tables:

The best way forward is to download the latest driver for your card, and the two immediately prior versions. Also download the Display Driver Uninstaller DDU. Install each of those three driver versions in turn, using DDU each time to completely remove the previous driver version. If you still see this BSDO with all three driver versions then it's almost certainly a hardware issue with the graphics card .

Hi Ubuysa,

I have read your response thoroughly, and found myself relieved that the NVIDIA driver is confirmed to be at fault here. Although it has been 3 years ever since I first used my laptop, I still believe that my GPU card on it can still last me for a few more years to come, so the hardware issue may not actually be a thing, at least for now, because I recently did reapply thermal paste and cleaned my laptop internally as well to make sure it's not overheated even during heavy use.

Anyway, following the solution you provided at the end of your response, I have reinstalled all 3 driver versions of NVIDIA graphics driver, and am currently settling on the latest driver version. I am monitoring my system during the moment I use it and will keep you updated later with how things turn out, which will be probably in a few weeks from now.

Ok. Keep us posted.

ubuysa said:

Ok. Keep us posted.

Hi ubuysa,

It has been more than a month, and so far so good. I have not encountered any BSODs with NVIDIA's driver, so I can safely say that your solution has helped me out a lot.
Thank you so much for the help!

I'm happy to have been able to help.