BSOD: hal.dll, ntoskrnl.exe

Feb 5, 2019
10
0
10
Hello everyone,

We build a PC this Monday and today I had it run Witcher 3 in order to see if it behaves normally when running a game.
And after about 3 hours, I got a BSOD while tabbing out of Witcher 3 after an unusual amount of time passed loading a cutscene. I got a "dpc watchdog violation" code and after loading the .dmp it gives me the following:
hal.dll+17a9
ntoskrnl.exe+2651d5

I did some generic troubleshooting as dictated by Google:
-No unrecognized device in device manager
-Checked if SATA/AHCI drivers are up to date (they are)

Specs:
OS: Windows 10 Pro 17763
CPU: i9-9900k (stock)
Motherboard: Gigabyte Aorus Master Z390
RAM: Corsair Vengeance RGB PRO 2x8Gb @ 3200 Mhz (XMP on)
Boot Drive: Samsung Pro 970 512GB (M.2)

GPU: Gigabyte Gaming OC RTX 2080 Ti (oc'd:
->Clock: 1750~2000 Mhz (using msi afterburner OC-Scanner's curve, about +150 in average)
->Memory ~7300 (+300)
->Power Limit 122 % (max), when I ran Witcher3 without a frame cap (for about 20 minutes), GPU-Z would measure up to 370 Watts.
->Custom Fan Curve: At this speed it was at about 65~70 %
I was regularly checking hardwaremonitor because, as stated above, I wanted to see how the system was performing: the GPU was under about 75~82 % of load and did not reach 80°C. The FPS was capped to 90 using RivaTuner. Temps and load on other components were decent.

Other devices:
Toshiba EXT HDD 1 TB
MoBo's WLAN antenna

I ran several benchmarks and load applications to see how the PC performs under stress. It only crashed one time when I played around with the BIOS's CPU OC Settings and ran CineBench. BIOS reverted anything I changed back and I didn't touch it since.
Is there any way to at least narrow it down? I have a hunch it's because of the OC and I will try turning it off and check what happens.

Any advice is appreciated, thanks in advance.
 
Feb 5, 2019
10
0
10
I ran Windows Debugger on it and had it perform an analysis. It got me this: https://pastebin.com/82mbzBfS

Now, it pointed me to:
PROCESS_NAME: Corsair.Service.CpuIdRemote64.exe

It seems to be a subprocess of iCue, Corsair's LED software. I actually had it deactivated on auto start and never ran it but I still saw the very same process running just now. Removed the software and the process is not there anymore. I hope that really was the issue. Nonetheless, I now doubt it had anything to do with GPU OC but I'll leave it for about a week to see if there are no blue screens and if there are not, I might consider going back to oc'ing.
If someone else took a look at the .dmp or paste and check if I missed something, I would highly appreciate it. I've seen threads on this forum where people struggled endlessly against this error and I hope I can weed this out before it gets more annoying.