Question At my wits' end with ACCESS_VIOLATIONs

[freezewindow]

Hey there good people,

I am at my wits' end. I recently upgraded my PC to the point where I replaced most of the system, only keeping the PSU, GPU and a hard drive.

Ever since that upgrade the PC has been unstable. I get BSODs on a semi-regular basis, sometimes twice a day, sometimes with 2 weeks between them, but they do happen alot. Different programs I run also crash, most of the time showing me some variation on the "ACCESS_VIOLATION" error, this happens to chrome in the middle of watching a youtube video, seemingly at random, but it also happens to games or the software I use for work (DAW and video editing softwares mostly).

I have made many attempts at diagnosing the problem. so far I have:

• wiped all drives and did a completely fresh install of windows 10 (I did this several times)
• took out one stick of RAM to run single channel, then did the same again with my secons stick of RAM
• updated the motherboards BIOS to the newest version.
• switched out the GPU for one I know to be stable
• reset BIOS settings several times
• ran Windows memory diagnostics several times -no errors
• ran memtest86 several times - no errors
• ran chkdsk several times - no errors
• updated all drivers I could think of that could be relevant to the issue

nothing spat out any useable info, no hardware changes I made helped at all. I don't really have a way to "evoke" these errors, they just happen, even if i leave the system alone. One of my recent BSODs happened at 3 in the morning while i was sleeping.

my current system looks as follows:

• Win10 Home 64-bit
• ryzen 5 3600x running stock
• MSI b450 GAMING PRO CARBON AC
• corsair vengeance LPX DDR4 3200MHz c16 2x16GB (ran both at base and using the XMP profile)
• GTX1060(3gb), now GTX1070 (no change between GPUs, other than the extra performance before a crash)
• Random1TBHDD with unreadable label
• Samsung 970 EVO 1TB
• Sandisk sata SSD 240GB

Here are the Dumps since my most recent reinstall:

https://drive.google.com/drive/folders/1-yNKLlDuniGl_c8UAIjTQZEBGaAgrM7F?usp=sharing

I hope you guys have some ideas, thanks in advance for reading all this stuff.

 

[rgd1101]

latest bios?
what is on the ssd/ hdd? remove the none os disk?

 

[freezewindow]

latest bios?
what is on the ssd/ hdd? remove the none os disk?

The board came 2 or 3 versions behind, current bios is 7B85v1C , which is the most current version.

I have not actually tried running the system without the other 2 drives (OS is on the samsung EVO), since I assumed those were unlikely culprits, but i'll try that tomorrow.

 

[freezewindow]

I have removed all drives but the samsung EVO and it took only about half an hour for another crash of chrome, and another 10 minutes for a crash of my editing software with "ACCESS_VIOLATION" as the error message given. I really have no idea what i can still do here.
could it be an issue with the motherboard? maybe the CPU? how would I tell?

 

[rgd1101]

what the cpu temp?
make and model of the psu?

 

[freezewindow]

what the cpu temp?
make and model of the psu?

The 3600X goes up to 82°C under heavy loads, but usually idles much lower.
The PSU is an XFX XT 500w 80+ Bronze, I know it's a bit underpowered, but it should be able to handle a low end 1060 and a 3600x just fine, shouldn't it?

Is what i'm experiencing likely a power issue ?

 

[johnbl]

Hey there good people,

I am at my wits' end. I recently upgraded my PC to the point where I replaced most of the system, only keeping the PSU, GPU and a hard drive.

Ever since that upgrade the PC has been unstable. I get BSODs on a semi-regular basis, sometimes twice a day, sometimes with 2 weeks between them, but they do happen alot. Different programs I run also crash, most of the time showing me some variation on the "ACCESS_VIOLATION" error, this happens to chrome in the middle of watching a youtube video, seemingly at random, but it also happens to games or the software I use for work (DAW and video editing softwares mostly).

I have made many attempts at diagnosing the problem. so far I have:

• wiped all drives and did a completely fresh install of windows 10 (I did this several times)
• took out one stick of RAM to run single channel, then did the same again with my secons stick of RAM
• updated the motherboards BIOS to the newest version.
• switched out the GPU for one I know to be stable
• reset BIOS settings several times
• ran Windows memory diagnostics several times -no errors
• ran memtest86 several times - no errors
• ran chkdsk several times - no errors
• updated all drivers I could think of that could be relevant to the issue

nothing spat out any useable info, no hardware changes I made helped at all. I don't really have a way to "evoke" these errors, they just happen, even if i leave the system alone. One of my recent BSODs happened at 3 in the morning while i was sleeping.

my current system looks as follows:

• Win10 Home 64-bit
• ryzen 5 3600x running stock
• MSI b450 GAMING PRO CARBON AC
• corsair vengeance LPX DDR4 3200MHz c16 2x16GB (ran both at base and using the XMP profile)
• GTX1060(3gb), now GTX1070 (no change between GPUs, other than the extra performance before a crash)
• Random1TBHDD with unreadable label
• Samsung 970 EVO 1TB
• Sandisk sata SSD 240GB

Here are the Dumps since my most recent reinstall:

https://drive.google.com/drive/folders/1-yNKLlDuniGl_c8UAIjTQZEBGaAgrM7F?usp=sharing

I hope you guys have some ideas, thanks in advance for reading all this stuff.

i looked at one of the memory dumps. it looked like your graphics driver was involved in the bugcheck.
I would update your nvidia driver to the current version.
It looks like the microsoft video streaming was having issues. most likely due to your tap expressvpn software
but that is just a guess. system was running for over 3 days 18 hours and the direct x was passed a bad address (zero)

 

[freezewindow]

i looked at one of the memory dumps. it looked like your graphics driver was involved in the bugcheck.
I would update your nvidia driver to the current version.
It looks like the microsoft video streaming was having issues. most likely due to your tap expressvpn software
but that is just a guess. system was running for over 3 days 18 hours and the direct x was passed a bad address (zero)

hey, thanks for taking a look!

I assumed the driver was a problem as well, sadly this happened across several different driver versions, including the most current one. Crashes also happen without any VPN products installed.
anything else i can take a look at ? At this point i'm almost willing to just start replacing hardware, although with this system that'll cost me about 200 bucks a pop, so i'd prefer exhausting all other options first.

 

[johnbl]

hey, thanks for taking a look!

I assumed the driver was a problem as well, sadly this happened across several different driver versions, including the most current one. Crashes also happen without any VPN products installed.
anything else i can take a look at ? At this point i'm almost willing to just start replacing hardware, although with this system that'll cost me about 200 bucks a pop, so i'd prefer exhausting all other options first.

I looked at another one of your bugchecks:
chrome was running, the system went to close a file, then to a file filter handler, then closed the file and the file system went to clean up.
that took too long and the system bugchecked. I can not see the name of the driver. I would need to have a kernel memory dump to read out the info in the debugger.


(this could be chrome, or the graphics susbsystem, or the filter, or something in the storage subsystem)
from a minidump I can not read the list of file filters you have installed.

here are your internal error tracking: (but the logs are not saved in a minidump file)
4: kd> !wdfkd.wdfcrashdump loader
Retrieving crashdump loader information...
Local buffer 0x000001AFD05DA410, bufferSize 1960
----------------------------------------------
ImageName Version FxGlobals

Wdf01000 v1.29(0000)
acpiex v1.15(0000) 0xffffdd8f5f2e8d50
intelpep v1.15(0000) 0xffffdd8f5f2f4370
WindowsTrustedRT v1.15(0000) 0xffffdd8f5f2f2370
WindowsTrustedRTProxy v1.15(0000) 0xffffdd8f5f2ed370
msisadrv v1.15(0000) 0xffffdd8f6090fd40
vdrvroot v1.15(0000) 0xffffdd8f6087dcd0
EhStorClass v1.15(0000) 0xffffdd8f609abde0
amdpsp v1.11(0000) 0xffffdd8f6419dbb0
cdrom v1.15(0000) 0xffffdd8f5fa11df0
Vid v1.15(0000) 0xffffdd8f64384db0
CompositeBus v1.15(0000) 0xffffdd8f654dadb0
umbus v1.15(0000) 0xffffdd8f643dedb0
Ucx01000 v1.15(0000) 0xffffdd8f60821db0
USBXHCI v1.23(0000) 0xffffdd8f6429cdb0
wdiwifi v1.15(0000) 0xffffdd8f643c2d70
vwifibus v1.15(0000) 0xffffdd8f642e1d70
HDAudBus v1.25(0000) 0xffffdd8f65471dc0
GPIOClx0101 v1.15(0000) 0xffffdd8f64380db0
amdgpio2 v1.15(0000) 0xffffdd8f6565dde0
AmdPPM v1.15(0000) 0xffffdd8f65639050
amdgpio3 v1.11(0000) 0xffffdd8f64380b90
NdisVirtualBus v1.15(0000) 0xffffdd8f65669340
USBHUB3 v1.15(0000) 0xffffdd8f60819d00
Nahimic_Mirroring v1.15(0000) 0xffffdd8f65b6cdd0
ibtusb v1.15(0000) 0xffffdd8f6522aa30
BthLEEnum v1.15(0000) 0xffffdd8f6525f860
monitor v1.15(0000) 0xffffdd8f67a41de0
MsLldp v1.15(0000) 0xffffdd8f68d0ede0
WinQuic v1.15(0000) 0xffffdd8f5f085de0 * find out what this is WinQuic **
PEAUTH v1.15(0000) 0xffffdd8f6a1dede0
dc1-controller v1.11(0000) 0xffffdd8f6e22ede0
xinputhid v1.11(0000) 0xffffdd8f6c8cb7c0
WUDFWpdFs v1.15(0000) 0xffffdd8f5f5eade0
WpdUpFltr v1.15(0000) 0xffffdd8f6ebf64e0
----------------------------------------------
\SystemRoot\system32\DRIVERS\lvrs64.sys Mon Oct 22 19:11:24 2012
\SystemRoot\system32\DRIVERS\lvuvc64.sys Mon Oct 22 19:12:08 2012

another old driver (provide audio enhancements and background noise reductions )
might be causing problems with audio processing by a video card.
I would remove to see if it helps
------------
one bugcheck had a code that
device USB\VID_0BDA&PID_0409\8&101b62bf&0&1
the error was device not present or not working properly

you should look for the device usb video device?

 

[freezewindow]

I looked at another one of your bugchecks:
chrome was running, the system went to close a file, then to a file filter handler, then closed the file and the file system went to clean up.
that took too long and the system bugchecked. I can not see the name of the driver. I would need to have a kernel memory dump to read out the info in the debugger.


(this could be chrome, or the graphics susbsystem, or the filter, or something in the storage subsystem)
from a minidump I can not read the list of file filters you have installed.

here are your internal error tracking: (but the logs are not saved in a minidump file)
4: kd> !wdfkd.wdfcrashdump loader
Retrieving crashdump loader information...
Local buffer 0x000001AFD05DA410, bufferSize 1960
----------------------------------------------
ImageName Version FxGlobals

Wdf01000 v1.29(0000)
acpiex v1.15(0000) 0xffffdd8f5f2e8d50
intelpep v1.15(0000) 0xffffdd8f5f2f4370
WindowsTrustedRT v1.15(0000) 0xffffdd8f5f2f2370
WindowsTrustedRTProxy v1.15(0000) 0xffffdd8f5f2ed370
msisadrv v1.15(0000) 0xffffdd8f6090fd40
vdrvroot v1.15(0000) 0xffffdd8f6087dcd0
EhStorClass v1.15(0000) 0xffffdd8f609abde0
amdpsp v1.11(0000) 0xffffdd8f6419dbb0
cdrom v1.15(0000) 0xffffdd8f5fa11df0
Vid v1.15(0000) 0xffffdd8f64384db0
CompositeBus v1.15(0000) 0xffffdd8f654dadb0
umbus v1.15(0000) 0xffffdd8f643dedb0
Ucx01000 v1.15(0000) 0xffffdd8f60821db0
USBXHCI v1.23(0000) 0xffffdd8f6429cdb0
wdiwifi v1.15(0000) 0xffffdd8f643c2d70
vwifibus v1.15(0000) 0xffffdd8f642e1d70
HDAudBus v1.25(0000) 0xffffdd8f65471dc0
GPIOClx0101 v1.15(0000) 0xffffdd8f64380db0
amdgpio2 v1.15(0000) 0xffffdd8f6565dde0
AmdPPM v1.15(0000) 0xffffdd8f65639050
amdgpio3 v1.11(0000) 0xffffdd8f64380b90
NdisVirtualBus v1.15(0000) 0xffffdd8f65669340
USBHUB3 v1.15(0000) 0xffffdd8f60819d00
Nahimic_Mirroring v1.15(0000) 0xffffdd8f65b6cdd0
ibtusb v1.15(0000) 0xffffdd8f6522aa30
BthLEEnum v1.15(0000) 0xffffdd8f6525f860
monitor v1.15(0000) 0xffffdd8f67a41de0
MsLldp v1.15(0000) 0xffffdd8f68d0ede0
WinQuic v1.15(0000) 0xffffdd8f5f085de0
PEAUTH v1.15(0000) 0xffffdd8f6a1dede0
dc1-controller v1.11(0000) 0xffffdd8f6e22ede0
xinputhid v1.11(0000) 0xffffdd8f6c8cb7c0
WUDFWpdFs v1.15(0000) 0xffffdd8f5f5eade0
WpdUpFltr v1.15(0000) 0xffffdd8f6ebf64e0
----------------------------------------------

Thank god you're here, making sense of that kind of stuff is beyond me.
Is there anything else I can do to aid the diagnosis ?

 

[johnbl]

Thank god you're here, making sense of that kind of stuff is beyond me.
Is there anything else I can do to aid the diagnosis ?

you can uninstall stuff, and maybe use
microsoft autoruns Autoruns for Windows - Windows Sysinternals | Microsoft Docs
to remove or disable suspect drivers ie uncheck a driver and reboot so it will be skipped while windows loads.
or delete the driver entry so it never loads again until the program is reinstalled.

only other thing you can do would be to change the memory dump type from a minidump to a kernel dump.
this will make a larger dump file located at c:\windows\memory.dmp and it will not strip out the internal logs, and will have info on the USB subsystems and the PLUG and play subsystem
as well as what was running on all of the other CPU cores at the time of the crash.

be sure to uninstall what ever logictech sound drivers you have installed (just because of the AGE of the drivers) bad sound drivers can mess with your motherboard sound and the sound support for your GPU that runs over your HDMI cable (even if you don't actually use it)

 

[freezewindow]

you can uninstall stuff, and maybe use
microsoft autoruns Autoruns for Windows - Windows Sysinternals | Microsoft Docs
to remove or disable suspect drivers ie uncheck a driver and reboot so it will be skipped while windows loads.
or delete the driver entry so it never loads again until the program is reinstalled.

only other thing you can do would be to change the memory dump type from a minidump to a kernel dump.
this will make a larger dump file located at c:\windows\memory.dmp and it will not strip out the internal logs, and will have info on the USB subsystems and the PLUG and play subsystem
as well as what was running on all of the other CPU cores at the time of the crash.

be sure to uninstall what ever logictech sound drivers you have installed (just because of the AGE of the drivers) bad sound drivers can mess with your motherboard sound and the sound support for your GPU that runs over your HDMI cable (even if you don't actually use it)

thank you for the help.

I have now changed the memory dump type to a kernel dump. I also used the software you linked to disable the logitech sound driver (and a hand full of others).

I am currently waiting for a program to crash or another BSOD to add usable information to this thread.

 

[freezewindow]

Alright, so I disabled that old sound driver, and the system seemed completely stable for quite a while. Sadly I did just get another crash with an "ACCESS_VIOLATION" message, it was WoW this time.

the crashlog reads something like this

Spoiler: Part of the crashlog

This application has encountered a critical error:

ERROR #132 (0x85100084) Fatal exception!

Program: C:\Users\freez\Desktop\save\World of Warcraft\retail\Wow.exe
ProcessID: 2232
ThreadID: 8304
Exception: ACCESS_VIOLATION

The instruction at "0x00007ffda1f4f0d0" referenced memory at "0x00000000000011f9".
The memory could not be "read".

<Exception.IssueType> Exception
<ExceptionType> Crash
<Exception.Summary:>
ACCESS_VIOLATION
(DBG-OPTIONS<FunctionsOnly SingleLine> DBG-ADDR<00007ffda1f48c36>("nvwgf2umx.dll") <- DBG-ADDR<00007ff785d64f5c>("Wow.exe") <- DBG-ADDR<00007ff78697437d>("Wow.exe") DBG-OPTIONS<>)
The instruction at "0x00007ffda1f4f0d0" referenced memory at "0x00000000000011f9".
The memory could not be "read".
<:Exception.Summary>
<Exception.Assertion:>
DBG-ADDR<00007ffda1f48c36>("nvwgf2umx.dll")
DBG-ADDR<00007ff785d64f5c>("Wow.exe")
DBG-ADDR<00007ff78697437d>("Wow.exe")
DBG-ADDR<00007ff7869762f6>("Wow.exe")
DBG-ADDR<00007ff786973db7>("Wow.exe")
DBG-ADDR<00007ff786976a95>("Wow.exe")
DBG-ADDR<00007ff787557acb>("Wow.exe")
DBG-ADDR<00007ff786d6fdd3>("Wow.exe")
DBG-ADDR<00007ff786d7027a>("Wow.exe")
DBG-ADDR<00007ff785c87f0b>("Wow.exe")
DBG-ADDR<00007ff785ce1309>("Wow.exe")
DBG-ADDR<00007ff785e17c38>("Wow.exe")
<:Exception.Assertion>
<Exception.HashBlock:>
DBG-OPTIONS<NoImage NoAddress NoFileLine NoFuncOffset NoDbgAddr>
ERROR #132
DBG-ADDR<00007ffda1f48c36>("nvwgf2umx.dll")
DBG-ADDR<00007ff785d64f5c>("Wow.exe")
DBG-ADDR<00007ff78697437d>("Wow.exe")
DBG-ADDR<00007ff7869762f6>("Wow.exe")
DBG-ADDR<00007ff786973db7>("Wow.exe")
DBG-ADDR<00007ff786976a95>("Wow.exe")
DBG-OPTIONS<>
<:Exception.HashBlock>
<Version> 9.0.2.36839
<Config> Retail
Revision: 1094529
<Jira.JiraProjectId> 11106
<Jira.JiraProjectKey> WOW9
<Exception.ProjectId> 10
<Exception.BuildNumber> 36839
<Exception.Branch> 9.0.2
<WowProject> Mainline
<Application> World of Warcraft
<ApplicationType> Client
Type: WoW
<Executable> Wow.exe
Executable UUID: 911FDFFD-C044-4D0A-9619-54003C928398
<Wow.Platform> X64
<Exception.Platform> Win 10 (64-bit)
User: freez
Computer: DESKTOP-AOR5H0J
Virtual Memory: 0.00 MB
Free Disk Space: 264.48 GB
Exe Built: Dec 3 2020 17:10:42
<InitFlags> 0000007F 00001FFF 00000400
App Up Time: 0 days, 0 hours, 2 minutes, 53 seconds
System Up Time: 0 days, 2 hours, 37 minutes, 10 seconds
<TimingMethod> QueryPerformanceCounter
Current Frame: 9017
Current Time: 9429240
Current Tick: 17
<SystemResumes> 0
<CharLogins> 1
Number of thread pool threads (Total): 27
<LauncherName> Battle.net.exe
<Locale.Text> enUS
<Locale.Audio> enUS

I added the plain text and the .dmp to the google drive folder, afaik the most personal info in there is probably the battletag.

https://drive.google.com/drive/folders/1-yNKLlDuniGl_c8UAIjTQZEBGaAgrM7F?usp=sharing


How likely is this to be something hardware related? Could it be worth it to buy some cheap processor that's compatible with this board for testing or something ? Is it likely to be a power issue? maybe a higher end PSU ?

 

[Sohom]

believe me its the ram... i too faced those issues, and it looks like ryzens are choosy when it comes to the ram... i too had issues with corsair lpx vengeance... u can get a different set of rams... i recently bought gskills 3200 mhz ram and found out they are actually samsung b-dies...

 

[AdamG]

No substantive suggestions for you, but it would be worth to check timings of ram to make sure you are running at the specified speed for that model. Not likely the reason, but just to be sure.

 

[freezewindow]

believe me its the ram... i too faced those issues, and it looks like ryzens are choosy when it comes to the ram... i too had issues with corsair lpx vengeance... u can get a different set of rams... i recently bought gskills 3200 mhz ram and found out they are actually samsung b-dies...

Did you also use stuff like memtest86 to check the ram? I have run these two sticks of RAM, both together and individually, through diagnostic tools countless times now and not once did I get an error of any kind with those

No substantive suggestions for you, but it would be worth to check timings of ram to make sure you are running at the specified speed for that model. Not likely the reason, but just to be sure.

If you're talking speeds, I tried running the RAM at the 3200mhz it's rated for, and I also tried to have it run with the default 2133 set by the motherboard, made no difference.
If you're talking actual timings, wouldn't that make the system completely unusable, if those were off ?

 

[Sohom]

Did you also use stuff like memtest86 to check the ram? I have run these two sticks of RAM, both together and individually, through diagnostic tools countless times now and not once did I get an error of any kind with those

Yes i did, mine came back with no errors as well... also what u can do is try a fresh install of windows 10 and see if there are any driver issues...

 

[freezewindow]

Yes i did, mine came back with no errors as well...

that does sound promising! Any particular RAM kit you can recommend? something you have known to not cause issues?

 

[AdamG]

If you're talking actual timings, wouldn't that make the system completely unusable, if those were off ?

I'm talking actual timings CL/CAS, tRCD, tRP, tRAS, CMD, and there are scenarios where the timing is incorrect but will result in sporadic bluescreens, blackscreens, and corruption of downloads/files. Nothing specifically hard about verifying the timings are correct, AUTO settings can incorrectly set timings in specific cases.

Use cpu-z or something(bios), and check memory timings compared to the advertised timings on the memory you purchased to verify, and then disregard the suggestion if they match.

 

[freezewindow]

I'm talking actual timings CL/CAS, tRCD, tRP, tRAS, CMD, and there are scenarios where the timing is incorrect but will result in sporadic bluescreens, blackscreens, and corruption of downloads/files. Nothing specifically hard about verifying the timings are correct, AUTO settings can incorrectly set timings in specific cases.

Use cpu-z or something(bios), and check memory timings compared to the advertised timings on the memory you purchased to verify, and then disregard the suggestion if they match.

thank you for the suggestion, sadly they do match at 16-18-18-36. I'm a bit disappointed tbh, that would've been a cool fix.

 

[freezewindow]

Just had another BSOD, this time with the "SYSTEM_SERVICE_EXCEPTION" error attached. I uploaded the dump file to the same folder as before ( https://drive.google.com/drive/folders/1-yNKLlDuniGl_c8UAIjTQZEBGaAgrM7F?usp=sharing ) filename is 121920-7312-01.dmp

From what i could gather through google-fu it seems like this one is usually a driver or a RAM thing. guessing Sohom is correct and i gotta replace the RAM ?
I would love it if someone could take a look at this most recent dump though.

 

[johnbl]

looks like you were running wow, the system needed to read a file, the read request went thru a filter and went to the file system
but at the file system windows cache manager detected that it was already in cache so it tried to give the copy in cache to the memory manager
when the memory manager tried to put the pages in memory it detected memory corruption and shutdown the system.
the system was up for
System Uptime: 8 days 6:23:04.272

I think this is a timing related problem.
you might consider: disable lazy writes on your drive (guess it would be to uncheck write caching)
Enable or Disable Disk Write Caching in Windows 10 | Tutorials (tenforums.com)
this will not fix the problem but would change the timing windows where the problem would occur
( to fix the problem I would look for firmware updates to the SSD, bios, and any chipset support for the storage )

if you have a ssd you might consider
disable prefetch and superfetch
Windows 10: How to Disable/Enable Prefetch and Superfetch - WinBuzzer

(superfetch should not have an effect on this bugcheck, but for SSD do you really need to have windows load up all of your ram with programs in case you might want to run something days later?) disable prefetch might change the timing.

problems like this are timing related and hard to pin point.



4: kd> !sysinfo machineid
Machine ID Information [From Smbios 2.8, DMIVersion 0, Size=2536]
BiosMajorRelease = 5
BiosMinorRelease = 14
BiosVendor = American Megatrends Inc.
BiosVersion = 1.C0
BiosReleaseDate = 06/11/2020
SystemManufacturer = Micro-Star International Co., Ltd.
SystemProductName = MS-7B85
SystemFamily = To be filled by O.E.M.
SystemVersion = 1.0
SystemSKU = To be filled by O.E.M.
BaseBoardManufacturer = Micro-Star International Co., Ltd.
BaseBoardProduct = B450 GAMING PRO CARBON AC (MS-7B85)
BaseBoardVersion = 1.0
4: kd> !sysinfo cpuinfo
[CPU Information]
~MHz = REG_DWORD 3800
Component Information = REG_BINARY 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0
Configuration Data = REG_FULL_RESOURCE_DESCRIPTOR ff,ff,ff,ff,ff,ff,ff,ff,0,0,0,0,0,0,0,0
Identifier = REG_SZ AMD64 Family 23 Model 113 Stepping 0
ProcessorNameString = REG_SZ AMD Ryzen 5 3600X 6-Core Processor
Update Status = REG_DWORD 1
VendorIdentifier = REG_SZ AuthenticAMD

 

[freezewindow]

looks like you were running wow, the system needed to read a file, the read request went thru a filter and went to the file system
but at the file system windows cache manager detected that it was already in cache so it tried to give the copy in cache to the memory manager
when the memory manager tried to put the pages in memory it detected memory corruption and shutdown the system.
the system was up for
System Uptime: 8 days 6:23:04.272

I think this is a timing related problem.
you might consider: disable lazy writes on your drive (guess it would be to uncheck write caching)
Enable or Disable Disk Write Caching in Windows 10 | Tutorials (tenforums.com)
this will not fix the problem but would change the timing windows where the problem would occur
( to fix the problem I would look for firmware updates to the SSD, bios, and any chipset support for the storage )

if you have a ssd you might consider
disable prefetch and superfetch
Windows 10: How to Disable/Enable Prefetch and Superfetch - WinBuzzer

(superfetch should not have an effect on this bugcheck, but for SSD do you really need to have windows load up all of your ram with programs in case you might want to run something days later?) disable prefetch might change the timing.

problems like this are timing related and hard to pin point.



4: kd> !sysinfo machineid
Machine ID Information [From Smbios 2.8, DMIVersion 0, Size=2536]
BiosMajorRelease = 5
BiosMinorRelease = 14
BiosVendor = American Megatrends Inc.
BiosVersion = 1.C0
BiosReleaseDate = 06/11/2020
SystemManufacturer = Micro-Star International Co., Ltd.
SystemProductName = MS-7B85
SystemFamily = To be filled by O.E.M.
SystemVersion = 1.0
SystemSKU = To be filled by O.E.M.
BaseBoardManufacturer = Micro-Star International Co., Ltd.
BaseBoardProduct = B450 GAMING PRO CARBON AC (MS-7B85)
BaseBoardVersion = 1.0
4: kd> !sysinfo cpuinfo
[CPU Information]
~MHz = REG_DWORD 3800
Component Information = REG_BINARY 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0
Configuration Data = REG_FULL_RESOURCE_DESCRIPTOR ff,ff,ff,ff,ff,ff,ff,ff,0,0,0,0,0,0,0,0
Identifier = REG_SZ AMD64 Family 23 Model 113 Stepping 0
ProcessorNameString = REG_SZ AMD Ryzen 5 3600X 6-Core Processor
Update Status = REG_DWORD 1
VendorIdentifier = REG_SZ AuthenticAMD

I made the suggested adjustments and had either bad luck and no changes, or those adjustments made the crashes more frequent. With both settings changed any given program will crash after around 10 to 15 minutes.

I did order some replacement RAM (different chip manufacturer, I made sure) and will try that to see if the RAM was actually the Issue.

 

[johnbl]

I made the suggested adjustments and had either bad luck and no changes, or those adjustments made the crashes more frequent. With both settings changed any given program will crash after around 10 to 15 minutes.

I did order some replacement RAM (different chip manufacturer, I made sure) and will try that to see if the RAM was actually the Issue.

making the problem happen faster is progress. you should put up the memory dumps and I can look at the back trace and see if it still looks like a timing problem. normally, I think the windows memory manager cache would refer to the cache RAM inside of the CPU rather than ram external to the CPU (RAM). generally, the cpu memory controller would detect a error in CACHE RAM and the CPU would either flush the cache or call a bugcheck 0x124

at least with intel CPU, how data moves thru the various levels of the cache in the CPU is controlled by the voltages applied to certain pins on the cpu each voltage sets a frequency for each level of the cache. Get the voltage wrong then the data is sampled at the wrong rate. kind of like going on a highway at 60 MPH with your truck and running into a car that is going 35 mph. it can be a problem. I know how to find these with a intel cpu but not always with AMD CPUs.

anyway put up the new memory dumps if you have them. I can take a quick look to see what caused the failure and maybe we can figure a fix.

 

[freezewindow]

making the problem happen faster is progress. you should put up the memory dumps and I can look at the back trace and see if it still looks like a timing problem. normally, I think the windows memory manager cache would refer to the cache RAM inside of the CPU rather than ram external to the CPU (RAM). generally, the cpu memory controller would detect a error in CACHE RAM and the CPU would either flush the cache or call a bugcheck 0x124

at least with intel CPU, how data moves thru the various levels of the cache in the CPU is controlled by the voltages applied to certain pins on the cpu each voltage sets a frequency for each level of the cache. Get the voltage wrong then the data is sampled at the wrong rate. kind of like going on a highway at 60 MPH with your truck and running into a car that is going 35 mph. it can be a problem. I know how to find these with a intel cpu but not always with AMD CPUs.

anyway put up the new memory dumps if you have them. I can take a quick look to see what caused the failure and maybe we can figure a fix.

hey, sadly the only program in the lineup of stuff I use daily that actually creates dumps on crash is WoW. I added blizzards dump to the google drive, maybe that actually has usable info.
thank you for your time!