This error indicates a driver irq issue and a stack buffer overflow during Windows 11 BSOD.
This error indicates a driver irq issue and a stack buffer overflow during Windows 11 BSOD.
I'm struggling and seeking help with a problem I'm facing with my PC setup. My computer has been showing BSODs between 2:30 a.m. and 5:00 a.m. each night. It works perfectly during the day, serving as both a workstation and a PLEX server. I've tried to track down what might be causing these issues and what steps I took to fix them. I'm not sure if these problems are connected, but it could help.
About a couple of weeks ago, my desktop screens went dark. If this was due to a GPU failure, I bought a replacement card and installed it, but nothing changed. I reinstalled the previous GPU and started removing RAM sticks (Corsair, four 16GB). Pulling one of them restored my video signal. I reached out to Corsair, who sent me four new RAM sticks and installed them. For a short time, everything was functioning properly.
Later, while using my PLEX server software, I encountered BSODs when scanning new media. I saw brief crashes with messages like "SYSTEM_SERVICE_EXCEPTION" and "DRIVER_VERIFIER_IOMANAGER_VIOLATION". I spent about a week trying to figure out the cause. Then, one of my internal Seagate 16TB drives failed. I tried to recover the data but had no success.
This week, I installed BlueScreenView and checked crash logs, which pointed to a "ntoskrnl.exe" driver file.
Troubleshooting steps I took:
- Updated BIOS
- Clean uninstall and reinstalled GPU drivers twice
- Ran Windows Memory Diagnostic with no issues
- Applied a Windows update
- Installed all available drivers for the ROG STRIX X570-E gaming motherboard
- Added the latest AMD chipset drivers
- Performed a chkdsk scan on all drives
Recent system dump:
Minidump
Any additional advice or assistance would be greatly appreciated. Thank you ahead of time.
Sure, try running DISM/SFC SCannow and check if it resolves the issue. Also, remember to clear CMOS after updating the BIOS.
Sure, try running DISM/SFC SCannow and check if it resolves the issue. Also, remember to clear CMOS after updating the BIOS.
I made the minidumps accessible, sorry about that.
To be honest, I ended up exploring a tricky path trying to fix this issue. The Driver Verifier was one of the methods I came across online or in a video. Now my mind is confused, and I don’t remember why I turned it on.
Download Microsoft autoruns64.exe, execute it and turn off this driver:
pwdrvio.sys Mon Jun 15 18:43:45 2009
Partition Wizard Disk I/O driver
reboot and check if the issue disappears.
this is just a guess. I didn't find the real cause of the bug in the minidump. It seems to be a null pointer. The system was running for 7 hours before the bugcheck occurred.
Additional information comes from @Roland Of Gilead regarding the BSOD caused by an invalid instruction pointer. The failing instruction is displayed here.
Some register values might be uninitialized or incorrect.
The following registers are set:
rax=0000003b65cf9225, rbx=0000000000000000, rcx=0000000000000001,
rdx=000000000000108c, rsi=0000000000000000, rdi=0000000000000000,
rip=0000000000000000, rsp=ffffd60f445649c0, rbp=ffffd60f44564ab0.
r8 and r9 are assigned specific values, while r10 is zero.
The instruction pointer is set to 0xffffd60f44564830, but the RIP register does not hold a valid address.
This issue occurred when the processor exited idle mode, and I've observed that some AMD CPUs become unstable during transitions from low power idle to high power operation.
To verify, disable C-States in your BIOS settings for all processors. Most systems have a global toggle for this feature.
Turning off C-States prevents the BSODs if it's the root cause.
Keep in mind the only drawbacks are increased idle temperature and slightly higher power consumption when idle, which are manageable with a good cooler.