[SOLVED] Over the last month or so, I've been getting random BSODs ?

[Sou Suzumi]

Hello

I have the following system:

AMD Ryzen 5 4650G
Asus B550m-Plus
32 GB (4 x 8 GB) RAM Crucial Ballistix 3200MT/s (XMP)
512 Gb M.2 NVME Adata XPG
550W Super Flower 80 Plus Gold power supply

And in the last month or so, I've been having random BSODs that I can't find the reason of.

At first I've thought it might be something memory-related. I built this rig with 2x 8 GB RAM, and then bought another 2 sticks a while later, but I've tried running the system with only the old sticks and still got the BSODs. Running it with only the new sticks also did the same, and I've tried running memtest86 and it didn't detect any problem (4 passes only, that is what I'm allowed to do in the free version). Running the memory with XMP profile disabled also presented the same system crashes.

I've also thought it was something heat-related. It's summer here, and it's impossibly hot. I've also bought a second display, so I've thought that maybe the extra strain on the chipset was pushing it above the limit. But I've bought a fan kit to introduce some airflow, and in my testing it shows both CPU and chipset are running at a relatively cold 40ºC.

After having the presence of spirit to disable the "Restart automatically" checkbox, I was able to see the error messages.

Last night, the BSOD was a KMODE_EXCEPTION_ERROR_NOT_HANDLED, which, by my quick research, is a catch-all term for when the system has no idea what went wrong, so it wasn't very helpful.

Today, just a while before, I got a KERNEL_AUTO_BOOST_INVALID_LOCK_RELEASE. While looking for this error, I've found this old thread: https://forums.tomshardware.com/threads/bsod-kernel-auto-boost-invalid-lock-release-help.3119950/
Unfortunately, the person with the problem stoped answering, and the question remains unsolved. But I've seen someone mentioning that it might be a CPU error, and that worried me a bit.

Detecting hardware errors at this level is a bit over my current capabilities, I don't even have the tools to test it. My CPU is still under warranty with the store, but I'd need to have some way to prove to them that the completely random crashes I'm having and that I have absolutely no idea the cause are actually the CPU's fault.

I have a minidump folder, I guess it's holding the last 5 crashes. I put everything in a zip and hosted it here: https://drive.google.com/file/d/189CxMx4eKZ1jLh1TLk6MpWveSj6FPgkH/view?usp=sharing


Can someone help me here?

 

[Colif]

KERNEL_AUTO_BOOST_INVALID_LOCK_RELEASE

i think this one is lan drivers but I get a friend to convert dumps and he will reply later with a file I can use to read conversion.

 

[Sou Suzumi]

i think this one is lan drivers but I get a friend to convert dumps and he will reply later with a file I can use to read conversion.

To be perfectly honest, today I did a thing I was supposed to have done since the beginning, and updated my drivers.
I'm still testing. Since it BSODs completely randomly, I guess I'll wait to see if I can go 2 or 3 days without it happening to see if it solved the problem.

 

[gardenman]

Hi, how did you update your drivers? Hopefully manually, one at a time, after researching the hardware that you are using. Unfortunately, if you used an automatic driver updater, you may have made things worse.

I ran the dump files through the debugger and got the following information: https://jsfiddle.net/aq7gv5x9/show This link is for anyone wanting to help. You do not have to view it. It is safe to "run the fiddle" as the page asks.

Spoiler: Dump File Results

File information:	011522-7984-01.dmp (Jan 15 2022 - 02:12:29)
Bugcheck:	KMODE_EXCEPTION_NOT_HANDLED (1E)
Probably caused by:	amdppm.sys (Process running at time of crash: System)
Uptime:	0 Day(s), 17 Hour(s), 07 Min(s), and 26 Sec(s)

File information:	011522-7171-01.dmp (Jan 15 2022 - 15:59:04)
Bugcheck:	KERNEL_AUTO_BOOST_INVALID_LOCK_RELEASE (162)
Probably caused by:	win32kbase.sys (Process running at time of crash: firefox.exe)
Uptime:	0 Day(s), 5 Hour(s), 01 Min(s), and 40 Sec(s)

File information:	121721-5843-01.dmp (Dec 16 2021 - 22:35:57)
Bugcheck:	IRQL_NOT_LESS_OR_EQUAL (A)
Probably caused by:	hardware (Process running at time of crash: System)
Uptime:	0 Day(s), 0 Hour(s), 44 Min(s), and 05 Sec(s)

File information:	121721-5734-01.dmp (Dec 17 2021 - 03:13:21)
Bugcheck:	RESOURCE_NOT_OWNED (E3)
Probably caused by:	dxgkrnl.sys (Process running at time of crash: csrss.exe)
Uptime:	0 Day(s), 4 Hour(s), 37 Min(s), and 00 Sec(s)

File information:	121621-5625-01.dmp (Dec 16 2021 - 00:20:07)
Bugcheck:	IRQL_NOT_LESS_OR_EQUAL (A)
Probably caused by:	hardware (Process running at time of crash: Battle.net.exe)
Uptime:	0 Day(s), 0 Hour(s), 13 Min(s), and 55 Sec(s)

Comment: The GPU tweaking driver "iomap64.sys" was found on your system. (AI Suite or GPU Tweak 2 or Armoury Crate)

Possible Motherboard page: https://www.asus.com/Motherboards-Components/Motherboards/TUF-Gaming/TUF-GAMING-B550M-PLUS/
The ASUS site is too slow for me to determine if you have a BIOS update or not. You are using BIOS version 1202. If there is an update, wait for additional information before deciding to update or not. Important: Verify that I have linked to the correct motherboard. Updating your BIOS can be risky. Never try it when you might lose power (lightning storms, recent power outages, etc).

This information can be used by others to help you. Someone else will post with more information. Please wait for additional answers. Good luck.

 

[Sou Suzumi]

Hi, how did you update your drivers? Hopefully manually, one at a time, after researching the hardware that you are using. Unfortunately, if you used an automatic driver updater, you may have made things worse.

I updated it through Armoury Crate, it gets the drivers for my motherboard directly from Asus.

Also, yes, you identified the MB correctly, it's an Asus TUF B550M-Plus.

Those are the mainboard drivers installed right now:

Before updating, I was a full version behind in each of them.

 

[gardenman]

I updated it through Armoury Crate

Good, that's one of the few trusted ones (the ones from the actual hardware manufacturer). Lets hope the updates help. I'm sure Colif will have additional ideas.

 

[Colif]

well, trusted in that it should just offer drivers for the motherboard. but the utility choice is another subject.

funny, I was going to say that Firefox being victim of 2nd crash reinforced my idea it was lan drivers until I saw this
dxgkrnl!CCompositionToken::InitializeUpdates+0x1c4

Almost every time I see DirectX crash, its GPU drivers

May 19 2020

amdkmdag.sys

AMD Graphics driver

I would expect there are newer now? Are you running off the CPU?
I would update them - https://www.amd.com/en/support/prev...ries-desktop-processors/amd-ryzen-5-pro-4650g
updating chipset drivers might do same thing - https://www.amd.com/en/support/chipsets/amd-socket-am4/b550

crash 3 looks like GPU too
as does crash 4.
crash 1 & 5 not so obvious. but could all be same thing.

 

[Sou Suzumi]

I would expect there are newer now? Are you running off the CPU?
I would update them - https://www.amd.com/en/support/prev...ries-desktop-processors/amd-ryzen-5-pro-4650g
updating chipset drivers might do same thing - https://www.amd.com/en/support/chipsets/amd-socket-am4/b550

Yeah, I don't have a dedicated GPU, the 4650G is an APU, I'm using the integrated graphics.

I only updated the chipset and graphic drivers through Armoury Crate, and never installed the Radeon dedicated software with the GPU drivers, since this is just VEGA integrated graphics. Would it be worth it installing it? It kinda looks like bloat.

 

[Colif]

Its listed as software for your GPU so I think it would be safe to install
If I click on here - https://www.amd.com/en/support/kb/release-notes/rn-rad-win-21-12-1 - it shows your CPU is supported
AMD Ryzen™ PRO Processors with Radeon™ Vega Graphics

Asus haven't updated the Windows 10 drivers on motherboard site for 6 months, the ones on AMD site are newer. Motherbaord makers aren't the fastest to update drivers, often they want to change the launcher so its got their name in it.

 

[Sou Suzumi]

To be fair, since yesterday I didn't BSOD again, but I had some pretty lightweight games crashing, like Terraria and Factorio, so I guess the graphics driver is still not OK.

I installed the Radeon software with the newest driver today, and did the option the software gives to remove all old drivers before installing the new ones.



Now we wait...

 

[Sou Suzumi]

OK, update: after two days without any kind of BSOD, I just got one.

IRQL_NOT_LESS_OR_EQUAL

The only thing I was doing differently from the other days was running Bluestacks.

Here's the minidump link:

https://drive.google.com/file/d/1cWPizyknapLgqQEeoCUpos68Fv7Ly_rs/view?usp=sharing

Thanks for the help.

 

[gardenman]

I ran the dump file through the debugger and got the following information: https://jsfiddle.net/uymjsLqe/show This link is for anyone wanting to help. You do not have to view it. It is safe to "run the fiddle" as the page asks.

File information:	011922-6000-01.dmp (Jan 19 2022 - 14:03:34)
Bugcheck:	IRQL_NOT_LESS_OR_EQUAL (A)
Probably caused by:	hardware (Process running at time of crash: VoicemodDeskto)
Uptime:	0 Day(s), 7 Hour(s), 11 Min(s), and 34 Sec(s)

This information can be used by others to help you. Someone else will post with more information. Please wait for additional answers. Good luck.

 

[Colif]

VoicemodDesktop.exe

what Microphone are you using?

Oct 09 2014

CMUAC.sys

C-Media USB Audio Class driver

Blue Snowball?

error says its hardware but could just be drivers. Just guessing based on what was running at time of crash and what is likely to be involved.

 

[Sou Suzumi]

VoicemodDesktop.exe

what Microphone are you using?

Oct 09 2014

CMUAC.sys

C-Media USB Audio Class driver

Blue Snowball?

error says its hardware but could just be drivers. Just guessing based on what was running at time of crash and what is likely to be involved.

I use a Redragon Zeus Headset.
It's a "white label" of the same type of the HyperX Cloud II.

It never ocurred to me updating it's drivers, I'll go check if there are any.

 

[Colif]

There doesn't appear to be. I don't see it listed here - https://www.redragonzone.com/pages/download
is it USB?
if headset plugs into the motherboard via audio cables, it might use these

Sep 14 2021

RTKVHD64.sys

Realtek Audio System driver https://www.realtek.com/en/

they aren't so old

I Wonder what this is doing on here.

Dec 10 2020

igovsd.sys

MSI Sound Tune driver

https://www.techpowerup.com/267621/msi-releases-sound-tune-ai-powered-noise-cancellation-software
seems it only works on MSI laptops or desktops... so whats it doing on an Asus?

note: 99% of time i see DirectX its GPU, hence help above. That remaining 1% is sound, but I rarely see it. So its possible GPU wasn't cause above, its just the overwhelming conclusion most of the time.

 

[gardenman]

I Wonder what this is doing on here.
Dec 10 2020 igovsd.sys MSI Sound Tune driver

Apparently it's just a sound card driver and is available on other PCs too. I'm updating description to this:
Intelligo Virtual Audio Cable driver (Intelligo Technology)

 

[Colif]

Still worth asking about since error was possibly sound related.

 

[Sou Suzumi]

There doesn't appear to be. I don't see it listed here - https://www.redragonzone.com/pages/download
is it USB?
if headset plugs into the motherboard via audio cables, it might use these

Sep 14 2021

RTKVHD64.sys

Realtek Audio System driver https://www.realtek.com/en/

The headset can plug directly into the analog 3,5mm ports, but it also has a DAC, and I use it because it improves audio quality and gives me some controls directly on the cable.

 

[Sou Suzumi]

Hello

It just happened again, with around 4 hours of the system running.
At least now it's giving the same error, IRQL_NOT_LESS_OR_EQUAL, instead of random <Mod Edit> everytime like it used to do, so updating the MB drivers solved at least part of the issue.

I have the headset plugged in through USB mostly because I was too lazy to unplug it (and sometimes I need the mic), but today I was using a pair of IEMs plugged directly to the front panel 3,5mm jack, using the Realtek Audio driver as Windows default.

Interestingly enough, I've just noticed that after rebooting the Realtek Audio is not an option anymore in the Windows default device selection menu, it defaulted back to the USB audio device (the headset), and checking Device Manager it doesn't show the Realtek Audio device anymore. I'm going to reboot the system again to see if it comes back, I just didn't want to lose what I've already written.
So maybe it is an audio issue. Either way, I'm uploading the new minudump, in case it shines some new light on what's happening.

https://drive.google.com/file/d/1VdPor35mP0UIZeLGzeVnTbNOF8Tinx9X/view?usp=sharing

 

[Colif]

often bsod aren't just one thing.

I Wonder what this is doing on here.
Dec 10 2020igovsd.sysMSI Sound Tune driver

Apparently it's just a sound card driver and is available on other PCs too. I'm updating description to this:
Intelligo Virtual Audio Cable driver (Intelligo Technology)

so this might be the dac.

Virtual Audio Cable is a software product based on WDM multimedia driver that allows a user to transfer audio streams from one application to another. Any application is able to send an audio stream to the input side of a "virtual cable" while a corresponding application can receive this stream from the output side.

I don't know. i can see its used for streaming

its odd realtek isn't working randomly if you don't normally use it at all.

 

[Sou Suzumi]

its odd realtek isn't working randomly if you don't normally use it at all.

Well, in the last few days I've been using Realtek since I've got these new IEMs last Friday.

After rebooting the system, it started working again. Checking here, this is a Microsoft driver, version 6.0.9238.1 and dating from 09/14/2021.
Checking Asus website for this mb, they have an older driver, 6.0.8971.1 dating from 09/10/2020 for Windows 10, and they list the 6.0.9238.1 driver for Windows 11.

So I guess I'll download the older version from the website and downgrade my driver, and see what happens.

 

[gardenman]

I ran the dump file through the debugger and got the following information: https://jsfiddle.net/gsdpev1k/show This link is for anyone wanting to help. You do not have to view it. It is safe to "run the fiddle" as the page asks.

File information:	012122-8343-01.dmp (Jan 21 2022 - 11:59:24)
Bugcheck:	IRQL_NOT_LESS_OR_EQUAL (A)
Probably caused by:	ntkrnlmp.exe (Process running at time of crash: System)
Uptime:	0 Day(s), 3 Hour(s), 02 Min(s), and 06 Sec(s)

This information can be used by others to help you. Someone else will post with more information. Please wait for additional answers. Good luck.

 

[Sou Suzumi]

OK, so after downgrading the audio driver earlier today, something just happened.

While playing Terraria (a pretty light game that is not resource intensive), Windows crashed and rebooted. The weird thing is that it didn't BSOD, neither create a minidump file. It also wasn't a power surge, because the screen froze and the sound glitched out before restarting.
Realtek Audio was working right after the reboot, unlike the previous crash.

What I found weird was it not showing the BSOD or creating the minidump file. I tried looking at the settings to see if maybe some stealth Windows update hadn't changed it, but it's OK, "reboot automatically" is unchecked and it's telling the system to record a minidump.

This made me think that maybe my power supply is not being able to hold the system. But I use a Super Flower 550W 80plus Gold, and all calculators I used to decide the power supply tell me a 300W power supply is more than enough (the system with all bells and whistles at 100% load uses only 254W, since I don't have a dedicated GPU).


I think that during the weekend I'll save my data and try to format the system. Maybe even do the jump to Win 11, despite despising the new taskbar.

 

[Colif]

shutdown restart with no BSOD means windows wasn't involved. Likely have event 41(63) in event log, as windows would have found previous shutdown unexpected.
Sounds like hardware.

AMD Ryzen 5 4650G
Asus B550m-Plus
32 GB (4 x 8 GB) RAM Crucial Ballistix 3200MT/s (XMP)
512 Gb M.2 NVME Adata XPG
550W Super Flower 80 Plus Gold power supply

ram - i would expect you to get freezes, not a random restart
Often we blame GPU for restarts during games, you using the CPU
PSU - The Super Flower name alone was enough for me not to suspect it, but even good brands can have bad units, no one is perfect
NVME - Since you booted up afterwards, I don't think its nvme. If windows was completely gone and it wouldn't boot, I could point at nvme.

try running this on CPU - https://www.guru3d.com/files-details/prime95-download.html
Prime 95 how to Guide: http://www.playtool.com/pages/prime95/prime95.html

it checks CPU and Ram... since GPU is part of CPU, might as well start there.

Not as many tests available for PSU

1. the paper clip method - https://forums.tomshardware.com/threads/what-is-the-paperclip-method-of-testing-a-psu.1336402/
2. or multimeter,
3. or in the BIOS to check the +3.3V, +5V, and +12V. - https://www.lifewire.com/power-supply-voltage-tolerances-2624583

1 is mostly to check if it works
3 isn't overly useful as PC isn't under much load in BIOS
2 is best idea but not many people have a multimeter.


anything showing in reliability monitor?

 

[Sou Suzumi]

try running this on CPU - https://www.guru3d.com/files-details/prime95-download.html
Prime 95 how to Guide: http://www.playtool.com/pages/prime95/prime95.html

it checks CPU and Ram... since GPU is part of CPU, might as well start there.

OK, now I think we are getting somewhere.
See, I did what I've said I was gonna do, and reinstalled Windows last night. Actually, I went to the deep end and installed Windows 11 to check if it was not some driver incompatibility with Windows 10.

When I ran Prime 95, the system immediately crashed (blend option, the one that checks a bit of everything). After rebooting, I checked off the "automatic restart" option and told Windows to create a minidump file, and tried it again with small FFTs (maximum CPU stress). Again, immediately got a crash, but instead of immediately rebooting, the screen froze and I could see Prime 95 showing errors in some CPU cores (one was showing a rounding error and the other was showing some fatal errors). Then I got a BSOD and a minidump.

So this goes back to what I've said in my OP, that I've found that old forum post saying it might be a CPU problem, except now we have a way to consistently crash the system and see the CPU errors.

This is the minidump, btw, we can see if it shows anything different: https://drive.google.com/file/d/1ZCSjVL5dOveXzgkyNW7exBPaw1UHRijk/view?usp=sharing

Thanks for the help everyone.

Edit: I can't seem to reproduce the minidump crash anymore. I tried to do it to see if I could take a picture of the errors, but the system immediately restarts even with the option unchecked and Windows doesn't create a minidump. Did it 3 times, decided to stop before I end up completely killing the PC.
I just hope the store accepts the warranty claim with the proofs we got.