VIDEO_DXGKRNL_FATAL_ERROR BSOD on Windows 10

adwuga

Member
Joined
Apr 18, 2016
Posts
11
I have a new Sager NP8640, with an SSD that I installed windows 10 enterprise on. Whenever I use my graphics card for a few minutes, it BSODs. This includes just having GeForce Experience open, so it's not being over stressed (the temps are good under stress anyways).


BlueScreenView says it relates to dxgkrnl.sys and watchdog.sys. I don't have any antivirus software besides windows defender, so I'm not sure why watchdog.sys is involved.


Error code is 0x00000113


Things I've tried:
1- SFC /scannow


2- DISM.exe /Online /Cleanup-image /Scanhealth and /Restorehealth


3- Windows update


4- chkdsk /f


5- Uninstall and reinstall NVdia drivers using clean install, latest drivers, and drivers from Sager website. None of these fixed it.


6- Refresh PC, then reset PC


7- Driver verifier


8- Updated all my drivers




Here is the BSOD dump file, as well as dxdiag for any info on my machine that might be needed:
dump
dxdiag




Thanks for any help!




Edit: It crashed again but this time BlueScreenView also had ntoskrnl.exe.
Here is the dump:
dump
 
Hi adwuga,

I have an inkling this may be a conflict between your CPU's display drivers and the NVidia display drivers. Have you updated your Intel HD Graphics 530 drivers since you installed Windows 10? They look like they're a little old, you can get the latest ones for the 530 here.
 
Hmm, I thought I had updated them, but apparently not. The updated drivers didn't fix the problem though. I also tried the driver from the sager website but that didn't fix it either (it actually just broke windows more, but I can fix that part on my own).
 
Okay, since the issue is occurring when using your discrete graphics, I think the next step would be to test it with your integrated graphics disabled. Have you tried running in Safe Mode at all? You should create a restore point first, so that if after disabling your integrated graphics you're constantly crashing, you'll be able to load your restore point. You likely would also be able to enter into Safe Mode to re-enable them.

First, create a restore point:

  • Search for Create a restore point from the Start menu and select it from the list of results.
  • On the System Protection tab in System Properties, select Create.
8b41a921-f0b3-448c-8600-61594fb2753e.jpeg

  • Enter a description for the restore point, and then select Create > OK.
Then, open Device Manager and disable your 530's display drivers:

  • Search for Device Manager from the Start menu and select it from the list of results.
  • Under Display adapters, right-click on Intel(R) HD Graphics 530 and select Disable.
dm.jpg

Then just do your typical testing (e.g. run GeForce Experience for a bit) and see if it crashes.
 
Actually, looking at that last dump you attached where Bluescreenview blamed ntoskernel, the dump is reporting memory corruption. I'm assuming it's not a hardware issue, but you may have bad memory. You can follow the Memtest tutorial here on Sysnative, I don't know what the Sager warranty is like in regards to allowing you to move/remove components, but you can at least do the test on all modules. If Memtest does report errors, I would contact Sager with that information and have them either RMA your RAM, or your laptop.
 
Thanks for your detailed responses, Xer! Unfortunately, neither of them solved it. I ran memtest for a full 10 passes. No errors found (thank god).
When I disabled the intel display driver, the nvidia card didn't turn on, so I'm curious as to what was running the display, but when I ran an application with it, I still got a blue screen. Unfortunately I was away from the computer when it bluescreened, and bluescreenview shows the error message (which it calls bug check string) as blank, but it's still blaming the original two .sys files. I've attached the dump in case anything was different.
View attachment 042016-6781-01.zip
 
Thanks for trying the two troubleshooting tests. That's odd that the bug check string is blank in BSV, in WinDBG it's reported as [BUGCHECK_STR: 0x113]. Here's the stack text:
Code:
ffffd001`e8243768 fffff800`cfd43e1d : 00000000`00000113 00000000`00000019 00000000`00000002 00000000`000010de : nt!KeBugCheckEx
ffffd001`e8243770 fffff800`d0238717 : ffffe000`b43a5190 ffffffff`c000000e ffffe000`b43a5190 00000000`00000000 : watchdog!WdLogEvent5_WdCriticalError+0xcd
ffffd001`e82437b0 fffff800`d01e373b : ffffe000`b5d57900 ffffe000`b43a5100 ffffe000`b5d579c0 00000000`ff030000 : dxgkrnl! ?? ::NNGAKEGL::`string'+0x338a7
ffffd001`e8243840 fffff800`d27ab11d : 00000000`00000001 ffffd001`e82439d9 00000000`00000001 ffffe000`b43a5040 : dxgkrnl!DpiDispatchPower+0x8b
ffffd001`e8243960 00000000`00000001 : ffffd001`e82439d9 00000000`00000001 ffffe000`b43a5040 ffffd001`e8243a58 : nvlddmkm+0x16b11d
ffffd001`e8243968 ffffd001`e82439d9 : 00000000`00000001 ffffe000`b43a5040 ffffd001`e8243a58 ffffe000`00000004 : 0x1
ffffd001`e8243970 00000000`00000001 : ffffe000`b43a5040 ffffd001`e8243a58 ffffe000`00000004 00000000`00000000 : 0xffffd001`e82439d9
ffffd001`e8243978 ffffe000`b43a5040 : ffffd001`e8243a58 ffffe000`00000004 00000000`00000000 00000000`00000001 : 0x1
ffffd001`e8243980 ffffd001`e8243a58 : ffffe000`00000004 00000000`00000000 00000000`00000001 ffffe000`b4997000 : 0xffffe000`b43a5040
ffffd001`e8243988 ffffe000`00000004 : 00000000`00000000 00000000`00000001 ffffe000`b4997000 fffff802`55ccd4c5 : 0xffffd001`e8243a58
ffffd001`e8243990 00000000`00000000 : 00000000`00000001 ffffe000`b4997000 fffff802`55ccd4c5 ffffe000`b787b000 : 0xffffe000`00000004

Which NVidia drivers have you tried, just the current ones (364.72)? Depending on what you've tried, my next suggestion is going to be to do another clean uninstall of those drivers (personally I like to use DDU in Safe Mode), and then an installation of a slightly older version, 361.91. When you get to the "Options" step, choose "Custom (Advanced)" and un-check everything but the Graphics Driver.
 
I've tried 359.00 (from the sager website, the official drivers for this laptop) and 364.72, neither worked. I just did a clean install from the nvidia setup though, so I'll try using DDU in safe mode and report back (though I might have to wait till tomorrow, I'm rather busy).
 
Morning Guy's

there is something odd in that system and comparing the three dumps beside other information provided,
showed pretty differences.

the dump mentioned by xer with memory corruption shows
Code:
Built by: 10586.[COLOR=#ff0000]212[/COLOR].amd64fre.th2_release_sec.160328-1908

Seems Fast Ring win10 Track, right? :smile9:

That Dump 041716-7953-01.dmp also told us this:
Code:
CHKIMG_EXTENSION: !chkimg -lo 50 -d !pci
    fffff80016df4c30-fffff80016df4c3a  11 bytes - pci!PciBus_QueryDeviceRelations
    [ 40 57 41 56 48 83 ec 68:21 50 a0 5c 4d 00 00 00 ]
    fffff80016df4c3c-fffff80016df4c45  10 bytes - pci!PciBus_QueryDeviceRelations+c (+0x0c)
    [ 49 8b f8 4c 8b f1 0f 85:53 00 00 00 00 00 00 00 ]
    fffff80016df4c48-fffff80016df4c5c  21 bytes - pci!PciBus_QueryDeviceRelations+18 (+0x0c)
    [ ba 04 00 00 00 48 89 74:81 0c fe 16 71 02 00 00 ]
    fffff80016df4c5f-fffff80016df4c6d  15 bytes - pci!PciBus_QueryDeviceRelations+2f (+0x17)
    [ 48 8b cf e8 c1 0d fd ff:00 03 00 00 00 00 00 00 ]
    fffff80016df4c6f-fffff80016df4c75  7 bytes - pci!PciBus_QueryDeviceRelations+3f (+0x10)
    [ 48 85 c0 0f 85 65 db:00 a9 0c fe 16 71 02 ]
    fffff80016df4c78-fffff80016df4c86  15 bytes - pci!PciBus_QueryDeviceRelations+48 (+0x09)
    [ 8b 47 04 4c 89 6c 24 48:d9 51 bb f9 54 00 00 00 ]
    fffff80016df4c88-fffff80016df4ca4  29 bytes - pci!PciBus_QueryDeviceRelations+58 (+0x10)
    [ 00 48 89 6c 24 60 8d 50:03 00 00 00 53 00 00 00 ]
    fffff80016df4ca7-fffff80016df4cae  8 bytes - pci!PciBus_QueryDeviceRelations+77 (+0x1f)
    [ 45 33 c9 48 89 9c 24 80:00 21 50 a0 5c 4d 00 00 ]
    fffff80016df4cb0 - pci!PciBus_QueryDeviceRelations+80 (+0x09)
    [ 00:03 ]
    fffff80016df4cb2-fffff80016df4cb9  8 bytes - pci!PciBus_QueryDeviceRelations+82 (+0x02)
    [ 45 33 c0 48 c7 44 24 20:00 00 00 00 00 00 00 00 ]
    fffff80016df4cbc-fffff80016df4cc5  10 bytes - pci!PciBus_QueryDeviceRelations+8c (+0x0a)
    [ 00 00 33 d2 48 8d 8f 90:77 0c 00 00 f9 0c fe 16 ]
    fffff80016df4cc7-fffff80016df4ce5  31 bytes - pci!PciBus_QueryDeviceRelations+97 (+0x0b)
    [ ff 15 73 38 ff ff 48 8d:00 69 9a ad b7 ee 00 00 ]
    fffff80016df4ce7-fffff80016df4cfe  24 bytes - pci!PciBus_QueryDeviceRelations+b7 (+0x20)
    [ 33 d2 8b f0 ff 15 47 38:00 21 0d fe 16 71 02 00 ]
    fffff80016df4d00-fffff80016df4d1c  29 bytes - pci!PciBus_QueryDeviceRelations+d0 (+0x19)
    [ 00 48 8b 05 f8 d2 fe ff:03 00 00 00 00 00 00 00 ]
    fffff80016df4d1e-fffff80016df4d28  11 bytes - pci!PciBus_QueryDeviceRelations+ee (+0x1e)
    [ 0f 85 3e db 00 00 44 8b:00 00 21 50 a0 5c 4d 00 ]
    fffff80016df4d2c-fffff80016df4d2f  4 bytes - pci!PciBus_QueryDeviceRelations+fc (+0x0e)
    [ 48 85 ed 0f:ee 00 00 00 ]
234 errors : !pci (fffff80016df4c30-fffff80016df4d2f)

MODULE_NAME: memory_corruption

My guess, happened during swapping / paging into memory and an unknown Filterdriver may be responsible for this.

while the corresponding Eventlog says problems with the
Code:
Event[17]:
  Log Name: System
  Source: [COLOR=#ff0000]Service Control Manager[/COLOR]
  Date: 2016-04-18T18:29:13.696
  Event ID: 7023
  Task: N/A
  Level: Error
  Opcode: N/A
  Keyword: Classic
  User: N/A
  User Name: N/A
  Computer:{whiped out}
  Description: 
The User Data Access_4a47a service terminated with the following error: 
[COLOR=#ff0000]Unable to complete the requested operation because of either a catastrophic media failure or a data structure corruption on the disk[/COLOR].

each time the DB is opened, not a normal feature.

What is loaded into the System differs heavily depending on your started applications, it is just a snappshot from one CPU-Kernel, you have 8 of them
going with intels latest Generation Intel(R) Core(TM) i7-6700HQ and Chipset. :cool3:

As you also have Logitech Setpoint running on your System with the fameous KHALMNPR.exe (hooking on ntdll and others) including several DLL's which is worth for troubles since XP, i really fear for stability, get rid of or update to the latest Version.

Your latest Dump shows Built by: 10586.0.amd64fre.th2_release.151029-1700
what is going on here?

personaly i would likely see / have an full Kerneldump (memory.dmp) to see whats going on in the machine.

Several Networkingproblems are also present in your machine (not responsible for the BSOD) like that below, the cause may be KillerService.exe brought to you by Qualcomm
Code:
Event[13]:
  Log Name: System
  Source: Microsoft-Windows-DNS-Client
  Date: 2016-04-18T18:31:23.096
  Event ID: 8020
  Task: N/A
  Level: Warning
  Opcode: Info
  Keyword: N/A
  User: S-1-5-20
  User Name: NT AUTHORITY\NETWORK SERVICE
  Computer: {whiped}
  Description: 
The system failed to register host (A or AAAA) resource records (RRs) for network adapter
with settings:

           Adapter Name : {721082FE-9F93-4AC3-986C-2095FF0ECA28}
           Host Name : {whiped}
           Primary Domain Suffix : mit.edu
           DNS server list :
                 209.18.47.61, 209.18.47.62
           Sent update to server : <?>
           IP Address(es) :
             2604:2000:80cb:2100:713d:53e0:fbd5:44ca, 192.168.0.14

The reason the system could not register these RRs during the update request was because of a system problem. You can manually retry DNS registration of the network adapter and its settings by typing 'ipconfig /registerdns' at the command prompt. If problems still persist, contact your DNS server or network systems administrator. See event details for specific error code information.

regards
Michael
 
Morning, at least in Ger :-)

as the week wasn't weak i had not time enough and a short inspection of the memory.dmp
seems pretty the same as bevor.

Code:
 # Child-SP          RetAddr           Call Site
00 ffffd001`9263b768 fffff801`180d3e1d nt!KeBugCheckEx
01 ffffd001`9263b770 fffff801`18208717 watchdog!WdLogEvent5_WdCriticalError+0xcd
02 ffffd001`9263b7b0 fffff801`181b373b dxgkrnl! ?? ::NNGAKEGL::`string'+0x338a7
03 ffffd001`9263b840 fffff801`1ae9b11d dxgkrnl![COLOR=#ff0000]DpiDispatchPower+0x8b[/COLOR]
04 ffffd001`9263b960 fffff801`1ae9a45e nvlddmkm+0x16b11d
05 ffffd001`9263ba40 fffff803`50389dfa nvlddmkm+0x16a45e
06 ffffd001`9263ba70 fffff803`50282125 nt!PopIrpWorker+0x1f6
07 ffffd001`9263bb10 fffff803`503c0126 nt!PspSystemThreadStartup+0x41
08 ffffd001`9263bb60 00000000`00000000 nt!KiStartSystemThread+0x16

whenever the powerstate changes for the nVidia, the Card failed.
A very quick shot showed, intel-Grafik isn't involved and the Stacks overall looked common.

From that findings, the nVidia has a Problem, the Hardware itself and calling for warranty isn't the baddest idea.

currently i did not dig down the device and powerprofile for that card, so not proofed yet but very likely.

regards
Michael
 
So I'll have to send it back? I'll have to wait a month or so, I'm not in a good position right now to give up my laptop.
Is there any way to verify that it's a hardware problem?

Thanks so much for your help!
 
Hi adwuga,

I don't think the BIOS Sager uses has any built-in diagnostic tools. You could probably run a live-OS version of Linux Mint from a USB drive, and then run GpuTest 0.2.0 (the reason I say "probably" is you might need to install the NVidia Linux drivers for the GTX 970M, in which case a live boot won't work for this purpose). Windows 10 will let you do several varieties of an in-place repair/refresh/reset, but if you were having the problem right out-of-the-box, they're unlikely to correct the issue, even if it is only related to software.

If you haven't already, I would contact Sager and see if they have any diagnostic tests they'd want you to run. In some cases, they might even have their own kernel analysts who would look into the problem. Even if you're not ready to RMA yet, it would be best to get the ball rolling for when you are ready, as there may be several steps before they will RMA the unit.
 
Hi Xer, I'm having some trouble getting Linux Mint to work. It's installed properly and booting from the USB, but when the countdown finishes or I tell it it boot from USB in regular or compatability mode, the computer just restarts. I also tried Mint version 13, and that goes to a gray screen and then a black console screen with a bunch of text. I also tried Ubuntu, and it booted up fully after a few weird gray screens, but wouldn't accept mouse or keyboard input.
I've tried all the different varieties of windows' refresh/repair/reset with no luck :(
While in the BIOS, I saw an option under "Advanced Chipset" called "GPU Performance Scaling" which was enabled, and apparently allows overclocking. I disabled it and booted back into Windows, and when I used my graphics card it still BSOD'd, but gave VIDEO_TDR_FAILURE as the error instead. Here is the dump: View attachment 042716-7500-01.zip

I'll definitely contact Sager about RMAing it though, thanks for the advice!
 
Tested it again and it went back to the old VIDEO_DXGKRNL_FATAL_ERROR BSOD.
Not sure what's going on anymore.
 
I don't think the new minidump says much different unfortunately, it looks likely to be hardware failure. Since you plan on continuing to use the laptop for a bit before completing the RMA process, you will likely want to disable the display adapter either from Device Manager or the BIOS, just so you're not running into inadvertent BSoDs. If you still feel like poking around, you could try a live boot of OpenSUSE or Fedora and see if you fare any better - in Ubuntu, were you trying to use the laptop's built-in keyboard and trackpad, or an external USB mouse and keyboard?
 
Ah, that's too bad. Well I've contacted Sager about RMAing it.
I was using the built in keyboard and trackpad in Ubuntu. I'll try those other OS's, and if anything unexpected happens (like it not crashing), I'll report back. Otherwise, thank you guys so much for all the help! I would've had no idea what to do without you.
 

Has Sysnative Forums helped you? Please consider donating to help us support the site!

Back
Top