Black Screen Crashes and Reboots caused by FanControl and AMD ADLX Conflict (Logs Included)
Nobody has claimed this yet.
Assessment
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Newbie friendliness
- 25/100
Research direction
The report identifies FanControl's ADLX control path in the stack trace, especially ADLXControl.Reset() and ADLXWrapper.FanTuningStates.Reset(). Start by reproducing the crash with GPU fan control enabled and compare it with the reported read only workaround; the issue does not provide a repository file, test, or conclusive reproduction. Done would require identifying and fixing a reproducible cause, then confirming stable GPU fan control.
Written by the indexing model from the issue text.
Description
Hi everyone,
I've been experiencing random black screen crashes followed by a system reboot. During the crash, the screen goes black (sometimes with brief artifacts), but the PC case fans and RGB stay on until the system reboots. There is no BSOD. and the crashes can happen at any time even when GPU is idle.
- I am using a Sapphire Pulse RX 9060 XT (8GB)
- I ran some stress tests over GPU, CPU and my RAM. All passed. Also I did some game tests and benchmarks to utilize all of my devices to check for insufficient power. Everything worked fine and looks like there is no any hardware problem.
After a lot of testing, I narrowed it down to a conflict involving FanControl, AMD Adrenalin, and the ADLX sensor API.
Here is the timeline of my troubleshooting:
Initially, when adding the GPU to FanControl using LHM (Libre Hardware Monitor) or ADLX, it couldn't control the GPU fans and gave "Missing Speed and Control Sensors" errors. On my 5th try, ADLX finally worked, and I could control the GPU fans via FanControl. I set it to start on boot.
However, a day later, the system crashed.
Merged Event Viewer & Reliability Monitor Logs:
I checked the logs and found multiple crashes leading up to the unexpected shutdown. I have merged the repetitive logs for easier reading. Almost all of them share the 0xc0000005 (Access Violation) exception code.
1. AMD ADLX Service Crashes (Repeated 8+ times):
- Faulting application:
AMDADLXServ.exe - Faulting modules:
OLEAUT32.dllANDamdadlx64.dll - Exception code:
0xc0000005
2. FanControl Service Crash:
- Faulting application:
FanControl.Service.exe - Faulting module:
coreclr.dll - Exception code:
0xc0000005 - Stack Trace Snippet:
at ADLXWrapper.Bindings.ADLXPINVOKE.IADLXManualFanTuning_SetFanTuningStates(...) at ADLXWrapper.FanTuningStates.Reset() at FanControl.Domain.BackendProviders.ADLX.ADLXControl.Reset()
3. HWiNFO Crash (Used by my DeepCool case display):
- Faulting application:
HWiNFO64.EXE - Faulting module:
ntdll.dll - Exception code:
0xc0000005 - Note: The faulty application path recorded in reliability monitor is pointing to the HWiNFO app itself. Not the one which is implemented inside of my DeepCool case monitor program because it is using a kind of hwinfo thing inside of itself and does not need the HWiNFO app itself to be installed or running.
Other bizarre behaviors noticed before the crash:
- The GPU fans would randomly start spinning at idle, ignoring the Zero RPM mode set in both Adrenalin and FanControl.
- Once the FanControl system tray menu became completely unresponsive and had to be forcefully restarted.
- One time long ago when I was stress testing my GPU with Furmark and reading the temps, HWiNFO (the app itself not the one in DeepCool app) gave me an error/warning thing which said another app is trying to read the temps and can conflict with hwinfo. I dont remember the error completely and it was for when I didn't notice the crashes yet so I didn't pay enough attention to it and just closed my DeepCool monitor app because I got stressed and nothing went wrong during the test.
Current Workaround (6th Try):
I disabled software fan control for the GPU inside FanControl. Now, FanControl only reads the GPU temps to control my case fans, while AMD Adrenalin handles the actual GPU fan curve. Since doing this, the system seems stable.
I summarized the information, reports, and logs using Gemini and posted them here, as the original data was very lengthy and my testing hasn't reached a final, conclusive result yet.
I’ll share updates as they come and will eventually upload the complete report file. In the meantime, if you need any additional information for troubleshooting, or if anyone is facing a similar issue or has a workaround, please let me know in the comments below.
Also, please excuse any mistakes or formatting issues—I'm still quite new to GitHub!
Thanks!
Gemini's Analysis of the Logs:
The following is an AI analysis of the crash logs provided by Gemini to help pinpoint the root cause:
The logs clearly indicate a severe memory access violation (0xc0000005) caused by a race condition. When FanControl (via its ADLX Wrapper), HWiNFO, and AMD Adrenalin try to poll or control the GPU sensors simultaneously, it overloads the OLEAUT32.dll COM interface.
The fatal blow happens specifically when FanControl attempts to execute the SetFanTuningStates command to write new fan speeds to the GPU. This action causes AMDADLXServ.exe to crash. When the AMD ADLX service dies while interacting with the hardware, the GPU driver hangs, the display signal is lost (black screen), and Windows is forced to execute an emergency reboot (Event ID 41 / 6008).
The user's current workaround—setting FanControl to Read-Only mode for the GPU—is the correct solution, as it prevents FanControl from sending conflicting write commands to the unstable ADLX service.
Edit:
Worth noting that I closed the DeepCool app and its background service from Task Manager at startup, but the crashes still happened. However, when I uninstalled FC and was only running Adrenalin with the DeepCool monitor, there were no black screen crashes. (I'm not entirely sure about any service failures in Reliability Monitor during that time, but I believe there were none.)
Then I reinstalled a newer version of FC (V272), ended the DeepCool processes, and started testing, which gave the results you see above. This suggests there is a high chance that FC is interfering with Adrenalin, not DeepCool.
Before all of that I was using V271 and the black screen problem was still there.
And yes, I have .NET Framework installed.
- Dominant language
- No language data
- Stars
- 21.1k
- Forks
- 619
- Avg merge
- 11h 52m
- Merged PRs (30d)
- 1
Getting set up
This project ships no dev container, Dockerfile or contributing guide, so setting up is up to you: start from its README, and see our first-contribution guide for the general steps.
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
More from Rem0o/FanControl.Releases
-
bug next version
Difficulty 2/5 1-3 hours Newbie friendliness 68/100
Rem0o/FanControl.Releases#4220 · 1 comment ·
-
Difficulty 3/5 Half a day Newbie friendliness 35/100
Rem0o/FanControl.Releases#4228 · 1 comment ·
-
Difficulty 3/5 Half a day Newbie friendliness 55/100
Rem0o/FanControl.Releases#4225 ·
-
Difficulty 4/5 3-5 days Newbie friendliness 12/100
Rem0o/FanControl.Releases#4224 · 2 comments ·
-
Difficulty 4/5 3-5 days Newbie friendliness 18/100
Rem0o/FanControl.Releases#4222 · 14 comments ·
All issues in Rem0o/FanControl.Releases
Similar issues
-
Difficulty 2/5 1-3 hours Newbie friendliness 72/100
ok-oldking/ok-wuthering-waves#1785 ·
Maintainers usually reply within 1 day
-
Difficulty 2/5 1-3 hours Newbie friendliness 60/100
Maintainers usually reply within 1 day
-
Difficulty 2/5 1-3 hours Newbie friendliness 67/100
Maintainers usually reply within 1 day
-
BUG
Difficulty 2/5 1-3 hours Newbie friendliness 74/100
Maintainers usually reply within 1 day
-
BUG | `MCP 服务器` | 全局层行的停用/启用与删除报 404 server row not found,编辑保存会在 profile 层复制出重复行Possibly taken @w6242652 claimed this today. Open
Difficulty 2/5 1-3 hours Newbie friendliness 75/100