Windows 10: Partial workaround for Bugcheck 0x116, nvlddmkm.sys, VIDEO_TDR_ERROR

Partial workaround for Bugcheck code: 0x116 in module nvlddmkm.sys, VIDEO_TDR_ERROR

As many other people, I too faced this problem after upgrading my desktop system to Windows 10. Did some research and testing, now I'd like to share my knowledge.

In general, according to the web resources the problem is bound mostly to GeForce cards utilized under Windows 10 and manifests itself under two completely different conditions - under heavy load esp. playing games, or under lightweight or no load, for example web browsing. Fortunately I have no experience with the first case, my system is rock stable under load (FarCry 2, Stalker Clear Sky). On the other hand, it sometimes restarted itself without an evident cause, mostly not being used at all.

At this point I started experimenting, here are my observations.
The problem IS hardware-dependent. My notebook equipped with GeForce adapter too does not exhibit this behavior.
The problem IS OS-dependent. Tested W7 and W10 using the same W7 drivers, with completely different results.
The problem IS NOT dependent on graphic driver version number. Tested many versions of drivers, with consistent results.

Furthermore, I identified two conditions which may initiate the problem: while my computer goes into sleep state or while it finishes time-consuming operation - for example, copying ~120GB file in FreeCommander. Both conditions are related to the power saving mechanisms of PCs.

Finally I discover all settings needed to make my system as stable as it used to be for years while using Windows 7. Here they are.
Power Options - Plan:
- set Put the computer to sleep to Never
Power Options - Choose what the power button does:
- uncheck Show Sleep in Power menu
Power Options - Advanced Settings:
- set PCI Express Link State PM to Disabled
- set USB selective suspend to Disabled
........
(for Classic Shell installed only)
Classic Start Menu - Settings:
- fill Shutdown Menu Items with options "switch_user;logoff;lock;restart;hibernate" (excluding Sleep)

There is no need to disable power saving for processor, the problem is bound to PCI Express bus and devices. Also hibernation can be used without problems.

Tested on my desktop only, I cannot guarantee success for other systems, but you may give it a try. This workaround is suboptimal for notebooks too, I am aware of it. But I think a stable system is still more important than longer battery life.

In my opinion, the problem isn't caused by neither bad graphic driver nor faulty hardware, it lays in interaction between hardware (motherboard + graphic card), OS and drivers while servicing power savings. It only manifests itself in graphic driver nvlddmkm.sys (and only under Windows 10), the same driver itself works ok in older OS. To find a culprit, we have to look at the way how the new OS manages entering/exiting power states on some motherboards and graphic cards. Resolving it is job for Microsoft guys i'm afraid.

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

Conclusion: If you encounter 0x116 errors while your computer is/becomes lazy, at first you may try to completely disable all power saving options instead of fiddling with drivers.

:)

 

Varius Bsods, mostly when playing games

BugCheck 116, {ffff9d0a20d494a0, fffff80d32152a1c, ffffffffc000009a, 4}

Bug Check 0x116 VIDEO_TDR_ERROR | Microsoft Docs read the MS Document

Unable to load image \SystemRoot\system32\DRIVERS\nvlddmkm.sys, Win32 error 0n2

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

*** ERROR: Module load completed but symbols could not be loaded for nvlddmkm.sys

Probably caused by : nvlddmkm.sys ( nvlddmkm+9b2a1c )

Chipset Driver: amd-catalyst-15.7.1-sb-sata-ahci-win10-win8.1-win7.exe

Regards.

 

Windows 10 Keep Crashing (Freeze or BSOD)

Hi, would like to seek help having my system fixed. I keep screen either screen freezes or BSOD after an hour or using my PC. My latest mini dump displays this:

On Tue 8/8/2017 8:49:43 PM your computer crashed

crash dump file: C:\Windows\Minidump\080817-11015-01.dmp

This was probably caused by the following module: nvlddmkm.sys (nvlddmkm+0xA1F7D8)

Bugcheck code: 0x116 (0xFFFF910B33C774A0, 0xFFFFF80F28A6F7D8, 0xFFFFFFFFC000009A, 0x4)

Error: VIDEO_TDR_ERROR

file path: C:\Windows\System32\DriverStore\FileRepository\nv_dispi.inf_amd64_24ddebfb518b5a55\nvlddmkm.sys

product: NVIDIA Windows Kernel Mode Driver, Version 384.94


company: NVIDIA Corporation

description: NVIDIA Windows Kernel Mode Driver, Version 384.94

Bug check description: This indicates that an attempt to reset the display driver and recover from a timeout failed.

A third party driver was identified as the probable root cause of this system error. It is suggested you look for an update for the following driver: nvlddmkm.sys (NVIDIA Windows Kernel Mode Driver, Version 384.94 , NVIDIA Corporation).

Google query: NVIDIA Corporation VIDEO_TDR_ERROR

I've already installed the latest driver and rolled my PC back to 2 previous drivers than this one and it still happens. I'm running six GTX 1070s powered by two 1200w platinum PSUs. Motherboard is a Biostar TB250 BTC Pro, 16gb RAM and 128 SSD. I've replaced
the risers with new ones and it still happens. Sometimes just in a few minutes it happens, sometimes it would last an hour. I've already ran diagnostics for my RAM and did a chkdsc. Is there anything else I can do to fix this?

 

Neverending story...

After two weeks of enjoying rock-stable Windows 10 system, I decided to install Windows Server trial onto another disk. Then I rebooted into Windows 10 and... error 0x116 /or black screen/ is back. So I made restore from the image of stable Window 10 system - no luck. Tried all workarounds found here and elsewhere - still the same problem.

So let's analyze what was happened:
1. System Win10 is fully stable
- disk image created
2. Parallel OS installation
- independent, on another disk, hardware unchanged, no change of BIOS settings
3. System Win10 is unstable
- even if restored from good image.

The only thing that might cause this inconsistence is change of ACPI Config, probably done by Windows Server setup program. (ACPI includes power management and resource allocation for hardware components.) Unfortunately, my BIOS prevents changes in resource allocation. So I will study how to modify ACPI for working configuration, meanwhile I have to live with generic VGA driver instead of GeForce one. It is stable of course, but not useful for gaming and playing videos.

Is somebody here who has real knowledge in ACPI modding? I appreciate any help.

 

After trying many configuration changes to fully avoid 0x116 errors, I probably found a working solution - my system is rock-solid for last four days. Steps needed to make my system stable:

- downgrade geforce driver to version 340.76 (for win7/8)
- disable sleep mode, hibernation and other power savings in power options
- disable power savings of usb hubs and network adapters in device manager

The first step is essential. This driver is no longer accessible at nvidia site, I downloaded it from softpedia. In conjunction with other two steps which limit acpi functions, it did the trick.

The page where I found this info (thx junclj): How to fix Display Driver Stopped Responding and has Recovered How-tos
junclj Oct 23, 2015 @ 17:27:42
"The only way to solve display driver stopped responding is to rollback and install the old stable version such like “340.76”. I got an old computer with motherboard Asus P5KPL-AM. It is a 6 years old computer. The graphic card is GTX460 and my system is Windows 10. Since all the new graphic drivers have compatibility issues. I was rolled back from 358.50 to the old version 340.76. No more stopped responding after I installed the old stable version. Just stay at 340.76 for your old machine and turn off the update checker and notification."

Will do more tests in the next days to ensure that this solution is permanent, and if suppressing of power savings is really mandatory. If you encounter the same problem (esp. if you have older mobo with intel or nforce chipset) you may at least try that very old driver. It does not support directx 12, but I don't care.

Link to download driver:
Download NVIDIA OpenGL 4.5 Graphics Driver 340.76 Beta 64-bit for Windows Vista 64 bit, Windows 7 64 bit, Windows 8 64 bit

 

The final update

After last mentioned changes, my system has been almost stable, at least while browsing/video playing/gaming etc. The only problem was two or more simultaneous copying of large amount of data (mostly virtual machines) which occassionally ended in bugcheck 0x116.

Tried everything which I can find on the net or invent myself, with no success. Then I noticed that the crash occures only when copying data to/from external USB drive. The last thing I did was replacing drivers for USB3 add-on card - the ones from station-drivers.com instead of manufacturer supplied ones. And then, I have no one crash during last three weeks.

Conclusion: Nvidia Video_TDR_Error may be only manifestation of different problem, in this case bad drivers of other PCIe device. So not all video crashes are directly caused by bad graphics card/driver.