Question BSOD ntoskrnl.exe+3f71b0 multiple crashes need help pls

Jul 27, 2022
10
0
10
0
Pc Specs
CPU: AMD Ryzen 5 3600 6-Core

GPU: Nvidia RTX 1660 Super

MOBO: msi B450 Tomahawk MAX

RAM: corsair 16 gb

PSU: Corsair CX750M

For 2 weeks I get frequent BSODS, or computer just freezes, and its more frequent after i haven't used the pc in about a week (before the 2 weeks). It'll happen anytime playing games like fortnite, or browsing on google. Sometimes right even after i restart it. I downloaded bluescreenviewer and most of the BSODS come from the ntoskrnl.exe driver, or ntoskrnl.exe+3f71b0 to be more specific. Different errors come up every time, like memory_management, irql_not_equal, etc

I have done most of the solutions presented for the past few days. Did file integrity checks, memtested, ran a bunch of commands (chk dsk, sfc scannow) couldn't reset my pc so i resinstalled windows on it but it still didn't work. Updated most of my drivers like my gpu, cpu, etc. All hardware is plugged in. Updated drivers multiple times but I can't find the solution still. Anyone know what to do? I don't know if its a hardware problem or not

Here are some minidump files
https://1drv.ms/u/s!AvmvHsiIqhfdigGlledf9ePb6EBK?e=SERN9S
https://1drv.ms/u/s!AvmvHsiIqhfdiX_mJcwI8FgosNqR?e=JX7Zuw
 
first bugcheck:
access violation in cache manger/memory management code trying to free a page.

second bugcheck was also in memory management attempting to free up virtual memory

only thing I can suggest would be to turn off virtual memory to delete the pagefile.sys, reboot and turn it back on to create another pagefile.sys

the oldest bugcheck had 3 modified windows core files.
the newer bugcheck did not.
so the assumption would be that something made changes to the pagefile.sys that were not valid.
otherwise it kind of looks like a cpu bug or some low level problem with the cpu. Maybe a chipset bug. just not clear
 
Last edited:
Jul 27, 2022
10
0
10
0
first bugcheck:
access violation in cache manger/memory management code trying to free a page.

second bugcheck was also in memory management attempting to free up virtual memory

only thing I can suggest would be to turn off virtual memory to delete the pagefile.sys, reboot and turn it back on to create another pagefile.sys

the oldest bugcheck had 3 modified windows core files.
the newer bugcheck did not.
so the assumption would be that something made changes to the pagefile.sys that were not valid.
otherwise it kind of looks like a cpu bug or some low level problem with the cpu. Maybe a chipset bug. just not clear
hello, soo i actually did a bios update via flash drive to the most recent version (not the beta one), and my bootup time is a lot slower than normal, but the main problem still persists of bsods. I also changed my dram speed from 2133 back to 3200 mHz, which is what it was before. However, the bsods are not from ntoskrnl.exe+3f71b9 this time, but rather ntoskrnl.exe+3f6cf0. I also turned virtual memory off and on but i still get some bsods. Here are some new dump files
https://1drv.ms/u/s!AvmvHsiIqhfdigSAgeAPm4IbB7BJ?e=PTtECD
 
Last edited:
in this bugcheck some service host was running, (need kernel dump to find out which one)
the system tried to pre fetch some data before it was needed but got an access violation when it tried to read the page file name

- can not determine if the pfn is valid. you would have to provide a kernel memory dump not a minidump.

- the I can not tell if the kernel address is valid. it looks to be in the correct range (kernel range, not user range)

- debugger did not find any modified windows core files.

system was up 9 mins 16 seconds before crashing.
you do have one driver known to corrupt the system
ScpVBus.sys Sun May 5 14:31:26 2013
but it generally takes hours to start seeing any problems.
(driver was fixed in 2016 but it was a public driver and no one paid the $700 fees to release it with proper certificates)

that still leaves you with a cpu bug, chipset bug as a likely cause. You might also get this from incorrect voltages to the cpu pins (bios bug)

here are some of the amd drivers installed:
SystemRoot\System32\drivers\amdgpio2.sys Wed Mar 11 04:15:48 2020
\SystemRoot\System32\drivers\amdgpio3.sys Mon Mar 14 03:19:36 2016
SystemRoot\System32\drivers\AMDPCIDev.sys Fri Oct 2 02:01:46 2020
\SystemRoot\System32\drivers\amdpsp.sys Fri Mar 6 13:04:04 2020

The above drivers have dates on them so they are provided by the motherboard vendor or AMD
the two below have no dates so they were provided by Microsoft windows update
you might find that if you remove the above driver and directly install the AMD drivers directly from amd you might get proper updates.
You might also be able to directly tweak voltages in the cpu using the amdryzenmasterdriver.sys but you would need to talk to someone that does this and knows about your particular version of amd CPU. I would hope that you should get the system stable before attempting tweaking voltages.

with this particular cpu version I have seen many issues I just could not figure out and looked like CPU bugs. I have tried to find the CPU buglist but it is not released for this particular CPU
I have seen indications the cpu was released in china with issues and was resold on the grey market. reports of custom bios fixes released to the motherboard vendors but only for the Chinese websites. (data from Chinese forums read using google translate so the info may be pretty bad)

\SystemRoot\System32\drivers\amdppm.sys
\SystemRoot\system32\mcupdate_AuthenticAMD.dll

--------
11: kd> !sysinfo machineid
Machine ID Information [From Smbios 2.8, DMIVersion 0, Size=2594]
BiosMajorRelease = 5
BiosMinorRelease = 17
BiosVendor = American Megatrends International, LLC.
BiosVersion = 3.D0
BiosReleaseDate = 04/01/2022
SystemManufacturer = Micro-Star International Co., Ltd
SystemProductName = MS-7C02
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 TOMAHAWK MAX (MS-7C02)
BaseBoardVersion = 1.0
11: kd> !sysinfo cpuinfo
[CPU Information]
~MHz = REG_DWORD 3600
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 3600 6-Core Processor
Update Status = REG_DWORD 1
VendorIdentifier = REG_SZ AuthenticAMD
 
Jul 27, 2022
10
0
10
0
in this bugcheck some service host was running, (need kernel dump to find out which one)
the system tried to pre fetch some data before it was needed but got an access violation when it tried to read the page file name

- can not determine if the pfn is valid. you would have to provide a kernel memory dump not a minidump.

- the I can not tell if the kernel address is valid. it looks to be in the correct range (kernel range, not user range)

- debugger did not find any modified windows core files.

system was up 9 mins 16 seconds before crashing.
you do have one driver known to corrupt the system
ScpVBus.sys Sun May 5 14:31:26 2013
but it generally takes hours to start seeing any problems.
(driver was fixed in 2016 but it was a public driver and no one paid the $700 fees to release it with proper certificates)

that still leaves you with a cpu bug, chipset bug as a likely cause. You might also get this from incorrect voltages to the cpu pins (bios bug)

here are some of the amd drivers installed:
SystemRoot\System32\drivers\amdgpio2.sys Wed Mar 11 04:15:48 2020
\SystemRoot\System32\drivers\amdgpio3.sys Mon Mar 14 03:19:36 2016
SystemRoot\System32\drivers\AMDPCIDev.sys Fri Oct 2 02:01:46 2020
\SystemRoot\System32\drivers\amdpsp.sys Fri Mar 6 13:04:04 2020

The above drivers have dates on them so they are provided by the motherboard vendor or AMD
the two below have no dates so they were provided by Microsoft windows update
you might find that if you remove the above driver and directly install the AMD drivers directly from amd you might get proper updates.
You might also be able to directly tweak voltages in the cpu using the amdryzenmasterdriver.sys but you would need to talk to someone that does this and knows about your particular version of amd CPU. I would hope that you should get the system stable before attempting tweaking voltages.

with this particular cpu version I have seen many issues I just could not figure out and looked like CPU bugs. I have tried to find the CPU buglist but it is not released for this particular CPU
I have seen indications the cpu was released in china with issues and was resold on the grey market. reports of custom bios fixes released to the motherboard vendors but only for the Chinese websites. (data from Chinese forums read using google translate so the info may be pretty bad)

\SystemRoot\System32\drivers\amdppm.sys
\SystemRoot\system32\mcupdate_AuthenticAMD.dll

--------
11: kd> !sysinfo machineid
Machine ID Information [From Smbios 2.8, DMIVersion 0, Size=2594]
BiosMajorRelease = 5
BiosMinorRelease = 17
BiosVendor = American Megatrends International, LLC.
BiosVersion = 3.D0
BiosReleaseDate = 04/01/2022
SystemManufacturer = Micro-Star International Co., Ltd
SystemProductName = MS-7C02
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 TOMAHAWK MAX (MS-7C02)
BaseBoardVersion = 1.0
11: kd> !sysinfo cpuinfo
[CPU Information]
~MHz = REG_DWORD 3600
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 3600 6-Core Processor
Update Status = REG_DWORD 1
VendorIdentifier = REG_SZ AuthenticAMD

Hmmm ok I see, so i was not able to delete amdgpio2.sys or amdgpio3.sys files since they are being run in another program, not sure which. Heres a kernel memory dump file (hope i did it right, sorry if i didnt)
https://1drv.ms/u/s!AvmvHsiIqhfdigUl7uOTDY4LG2fy?e=0gNqzP

You said that "ScpVBus.sys is known to corrupt my system but it generally takes hours to start seeing any problems" Should i try deleting it ? Im not very knowledgeable w that kind of stuff, so i dont really know where to start and if it has any repercussions. But you said the main problem could be a cpu bug. Is it fixable, or is it more recommended to buy a new cpu since there wasnt a cpu buglist? I know how delicate those are, so im just afraid of messing up any parts if i keep using the same one. If i did install a new cpu, would i get less bsods? or would it be the motherboard thats the problem?
 
Last edited:
ScpVBus.sys can be removed using microsoft autoruns64 from here:
https://docs.microsoft.com/en-us/sysinternals/downloads/autoruns

you should go ahead and remove it just so it does not get blamed for these bugchecks.
driver is from scarlett crush and is for a game adapter.
code is public here:https://github.com/nefarius/ScpVBus/
--------
for the amd files you need to go to amd website and find the amd chipset uninstaller and run it and install the new version. follow the amd instructions. This has the best chance to fix the system.

--------
when I look at the system it is strange, all the bugchecks are related to cache management and prefetching by the cpu.
I can see their are 9 programs that want to run but they are blocked waiting for a free lock. The system looks like it is idle and is just trying to preload software for you. I would stop this by downloading microsoft rammap64.exe from here: https://docs.microsoft.com/en-us/sysinternals/downloads/rammap
run it as an admin and find the menu option for empty and select each empty menu item one at a time. The main hope it to clear the standby list and working set. This will not fix your problem but is likely to reduce the occurrence of the cpu prefetch errors.
(windows tracks everything you have run for months and during idle time it tries to guess what you might want to run and will preload programs into standby memory so if you try to run it, it will already be in memory and windows only has to mark the memory as active. I think this is what is having problems, maybe clearing the list will help)
clearing the list will also reduce the size of a full memory dump by the size of standby memory used (often by 5 to 8 GB)

cpu will always have a lot of bugs, generally these are found and documented then a patch is developed and sent to motherboard vendors. The motherboard vendors add the patch to the bios to override or disable the bad cpu function. microsoft also puts out a patch and loads it when windows loads. It does this since people to make the system more stable until people update their bios versions (if they do).
biggest problem is that AMD failed to document the cpu errors and the last list I could find was pretty old and was before this cpu was released. Intel has much better bug documentation for their cpus, chipsets and drivers.
it is nice to be able to relate a specific problem to an actual known bug and know the actual version number of a driver that fixes it. For example the last intel bug I found was already documented, and had a fix in a certain chipset driver version and all the person had to do was down load the driver directly from intel. The bug was in a chipset and would corrupt the pagefile.sys after several sleep/wake cycles until the system crashed. Nice to see the listed bug and listed fix and know it was fixed.
it seems AMD is hiding the buglist/fixlist now.
 
Last edited:
Jul 27, 2022
10
0
10
0
ScpVBus.sys can be removed using microsoft autoruns64 from here:
https://docs.microsoft.com/en-us/sysinternals/downloads/autoruns

you should go ahead and remove it just so it does not get blamed for these bugchecks.
driver is from scarlett crush and is for a game adapter.
code is public here:https://github.com/nefarius/ScpVBus/
--------
for the amd files you need to go to amd website and find the amd chipset uninstaller and run it and install the new version. follow the amd instructions. This has the best chance to fix the system.

--------
when I look at the system it is strange, all the bugchecks are related to cache management and prefetching by the cpu.
I can see their are 9 programs that want to run but they are blocked waiting for a free lock. The system looks like it is idle and is just trying to preload software for you. I would stop this by downloading microsoft rammap64.exe from here: https://docs.microsoft.com/en-us/sysinternals/downloads/rammap
run it as an admin and find the menu option for empty and select each empty menu item one at a time. The main hope it to clear the standby list and working set. This will not fix your problem but is likely to reduce the occurrence of the cpu prefetch errors.
(windows tracks everything you have run for months and during idle time it tries to guess what you might want to run and will preload programs into standby memory so if you try to run it, it will already be in memory and windows only has to mark the memory as active. I think this is what is having problems, maybe clearing the list will help)
clearing the list will also reduce the size of a full memory dump by the size of standby memory used (often by 5 to 8 GB)

cpu will always have a lot of bugs, generally these are found and documented then a patch is developed and sent to motherboard vendors. The motherboard vendors add the patch to the bios to override or disable the bad cpu function. microsoft also puts out a patch and loads it when windows loads. It does this since people to make the system more stable until people update their bios versions (if they do).
biggest problem is that AMD failed to document the cpu errors and the last list I could find was pretty old and was before this cpu was released. Intel has much better bug documentation for their cpus, chipsets and drivers.
it is nice to be able to relate a specific problem to an actual known bug and know the actual version number of a driver that fixes it. For example the last intel bug I found was already documented, and had a fix in a certain chipset driver version and all the person had to do was down load the driver directly from intel. The bug was in a chipset and would corrupt the pagefile.sys after several sleep/wake cycles until the system crashed. Nice to see the listed bug and listed fix and know it was fixed.
it seems AMD is hiding the buglist/fixlist now.
That's interesting, thank you for sharing that. I was able to do all of stuff mentioned above, however i did get a bsod right after.
heres another dump file, is there anything different?

https://1drv.ms/u/s!AvmvHsiIqhfdigaKlSUFEfiNPEtu?e=AbmrBf

Hmm so since amd is hiding the buglist, i gotta wait for my bios to release another update to overwrite this bug then?

https://1drv.ms/u/s!AvmvHsiIqhfdigc_drqr-ZkOnpjI?e=gHTiCl
 
Last edited:
That's interesting, thank you for sharing that. I was able to do all of stuff mentioned above, however i did get a bsod right after.
heres another dump file, is there anything different?

https://1drv.ms/u/s!AvmvHsiIqhfdigaKlSUFEfiNPEtu?e=AbmrBf

Hmm so since amd is hiding the buglist, i gotta wait for my bios to release another update to overwrite this bug then?

https://1drv.ms/u/s!AvmvHsiIqhfdigc_drqr-ZkOnpjI?e=gHTiCl
this was discord.exe running then memory mangement trying to insert a pagefile entry into a list and getting a bugcheck due to a bad pointer.

in previous bugchecks the last run program was discord,exe but the system reported it as idle and it looked like it was blocked from running.
maybe uninstall discord.exe and see if there is malware with it.

this is a better bugcheck that the previous ones. At least it is not in cpu prefetch or cache management. (those always seem to be timing related cpu issues)

the debugger does not report any corrupted windows core files.
Something tried to put a bad pagefile entry into a the pagefile.
since discord was the running process it would get the blame.
note: you now have AMDRyzenMasterDriver.sys Thu Jun 24 22:21:58 2021 installed. this is a overclock driver but can be used to fix various issues with the bios default values. You would need to talk to someone that knows about your specific cpu for what special setting to make. most likely someone on a AMD cpu hardware website.

I would not hold my breath to get a bios update for this cpu. I think the fixes were already put out to certain local markets in china.

where did you get this version of the cpu and how long have you had it? The last few people I tried to help with this cpu version just got a replaced the motherboard and when that did not work they got a different copy of the cpu. In one case it worked in another case it did not and they went and got a cpu upgrade (different newer version) and the problems went away.
your cpu info:
Identifier = REG_SZ AMD64 Family 23 Model 113 Stepping 0
ProcessorNameString = REG_SZ AMD Ryzen 5 3600 6-Core Processor
23 = 0x17
113 = 0x71

to find docs on this cpu you would google:
amd ryzen 5 3600 family 0x17 model 0x71 stepping version 0
looks like the cpu was released in 2018 so the list of bugs should be known.
https://www.cpu-world.com/CPUs/Zen/AMD-Ryzen 5 3600.html
old cpu version = less hope for a bios update to fix problems.

here is the most current buglist for this family of cpus.
it does not include your cpu.
Revision Guide for AMD Family 17h Models 30h-3Fh Processors
amd bugs are found as revisions
intel bugs are errata lists
 
Last edited:
Jul 27, 2022
10
0
10
0
this was discord.exe running then memory mangement trying to insert a pagefile entry into a list and getting a bugcheck due to a bad pointer.

in previous bugchecks the last run program was discord,exe but the system reported it as idle and it looked like it was blocked from running.
maybe uninstall discord.exe and see if there is malware with it.

this is a better bugcheck that the previous ones. At least it is not in cpu prefetch or cache management. (those always seem to be timing related cpu issues)

the debugger does not report any corrupted windows core files.
Something tried to put a bad pagefile entry into a the pagefile.
since discord was the running process it would get the blame.
note: you now have AMDRyzenMasterDriver.sys Thu Jun 24 22:21:58 2021 installed. this is a overclock driver but can be used to fix various issues with the bios default values. You would need to talk to someone that knows about your specific cpu for what special setting to make. most likely someone on a AMD cpu hardware website.

I would not hold my breath to get a bios update for this cpu. I think the fixes were already put out to certain local markets in china.

where did you get this version of the cpu and how long have you had it? The last few people I tried to help with this cpu version just got a replaced the motherboard and when that did not work they got a different copy of the cpu. In one case it worked in another case it did not and they went and got a cpu upgrade (different newer version) and the problems went away.
your cpu info:
Identifier = REG_SZ AMD64 Family 23 Model 113 Stepping 0
ProcessorNameString = REG_SZ AMD Ryzen 5 3600 6-Core Processor
23 = 0x17
113 = 0x71

to find docs on this cpu you would google:
amd ryzen 5 3600 family 0x17 model 0x71 stepping version 0
looks like the cpu was released in 2018 so the list of bugs should be known.
https://www.cpu-world.com/CPUs/Zen/AMD-Ryzen 5 3600.html
old cpu version = less hope for a bios update to fix problems.

here is the most current buglist for this family of cpus.
it does not include your cpu.
Revision Guide for AMD Family 17h Models 30h-3Fh Processors
amd bugs are found as revisions
intel bugs are errata lists
got it, will uninstall discord now
Ive had the Ryzen 5 3600 6-Core for about 2 years now and i bought it from walmart when it was brand new i believe. I never had any problems with it until like 2 weeks ago, and dunno if its important but the crashes started happening more often when i didnt use the computer for about a week. Before it would crash every few days but after it was like every day. Dont know if ill upgrade anytime soon, but i know for sure its not a hardware issue at least. Thank you by the way for helping so far
 
Last edited:
Jul 27, 2022
10
0
10
0
this was discord.exe running then memory mangement trying to insert a pagefile entry into a list and getting a bugcheck due to a bad pointer.

in previous bugchecks the last run program was discord,exe but the system reported it as idle and it looked like it was blocked from running.
maybe uninstall discord.exe and see if there is malware with it.

this is a better bugcheck that the previous ones. At least it is not in cpu prefetch or cache management. (those always seem to be timing related cpu issues)

the debugger does not report any corrupted windows core files.
Something tried to put a bad pagefile entry into a the pagefile.
since discord was the running process it would get the blame.
note: you now have AMDRyzenMasterDriver.sys Thu Jun 24 22:21:58 2021 installed. this is a overclock driver but can be used to fix various issues with the bios default values. You would need to talk to someone that knows about your specific cpu for what special setting to make. most likely someone on a AMD cpu hardware website.

I would not hold my breath to get a bios update for this cpu. I think the fixes were already put out to certain local markets in china.

where did you get this version of the cpu and how long have you had it? The last few people I tried to help with this cpu version just got a replaced the motherboard and when that did not work they got a different copy of the cpu. In one case it worked in another case it did not and they went and got a cpu upgrade (different newer version) and the problems went away.
your cpu info:
Identifier = REG_SZ AMD64 Family 23 Model 113 Stepping 0
ProcessorNameString = REG_SZ AMD Ryzen 5 3600 6-Core Processor
23 = 0x17
113 = 0x71

to find docs on this cpu you would google:
amd ryzen 5 3600 family 0x17 model 0x71 stepping version 0
looks like the cpu was released in 2018 so the list of bugs should be known.
https://www.cpu-world.com/CPUs/Zen/AMD-Ryzen 5 3600.html
old cpu version = less hope for a bios update to fix problems.

here is the most current buglist for this family of cpus.
it does not include your cpu.
Revision Guide for AMD Family 17h Models 30h-3Fh Processors
amd bugs are found as revisions
intel bugs are errata lists
So i deleted discord and got some bsods a few hrs later, here r some dump files got a massive wave of bsods
https://1drv.ms/u/s!AvmvHsiIqhfdigymGq7LMp4cSd0b?e=y3pAwV

https://1drv.ms/u/s!AvmvHsiIqhfdigg-uZpJUK4RM1SG?e=D6YLgG
https://1drv.ms/u/s!AvmvHsiIqhfdigqtZhTUPFUfZdzv?e=sdd7ir

maybe i might get a new processor lol
 
Last edited:
if the processor has been working for the last two years then the problem is more likely going to be in the updated support drivers. the cpus have a security processor that has been hacked. This caused the vendors to issue a bunch of required updates to the bios and the chipset support files. I would expect that they just screwed it up or you did not update something. I looked at the various amd files you now have installed, these look like they were updated from the version you used to have. your bios version is from
04/01/2022
biggest problem I see with that was most of the vendors did not push out the updated bios until after mid may. You might not have the updates. For the intel hack you have to run a special cpu firmware update to fix the microcode in the security processor. I would think the amd cpu would require something similar.

last bugcheck showed a memory management error while trying to validate a page file entry. address looked valid but I can not verify it. the call came out of a service host but I can not read the name from a minidump file.
second bugcheck was invalid (file size zero)

so far all of the bugs have been related to prefetching and fail when the page file entry is wrong. Any chance you have some storage that needs a firmware update? looks like you have the generic drivers and it would not have changed in the last few weeks.
the only thing that would have changed would be the microsoft amd cpu microcode patch that microsoft installs to fix known cpu bugs.
the file is: SystemRoot\system32\mcupdate_AuthenticAMD.dll.

I don't think I would get a new cpu yet, I think there just needs to be some bios fix or update.

none of the windows core files still seem ok, not corrupted.

maybe you should try turning off windows prefecthing. it could just be broken for some reason. here is instructions to stop it:
Enable or Disable SysMain (Superfetch) & Prefetch for SSD in Windows (thewindowsclub.com)

it might at least stop the bugchecks until you find a real fix.
i would disable prefetch and superfetch from the above link and see if it helps. Note; when I read about these it indicates that windows already disables the services on solid state drives. You might want to check its registry setting.
here is a sample trace of one of the bugchecks:
# Call Site
00 nt!KeBugCheckEx
01 nt!KiBugCheckDispatch
02 nt!KiPageFault
03 nt!MiIdentifyPfn
04 nt!MiIdentifyPfnWrapper
05 nt!PfpPfnPrioRequest
06 nt!PfQuerySuperfetchInformation
07 nt!ExpQuerySystemInformation
08 nt!NtQuerySystemInformation
09 nt!KiSystemServiceCopyEnd
0a 0x0

nt= nt kernel
Pfp = is the windows prefetch module
Mi= Memory manager Internal
Pfn= page file name.

so the calls are read from the bottom to the top.
a svchost.exe started the sequence of calls.
(need a kernel dump to figure out which service started the calls)

so basically, a service made some query, the kernel calls a superfetch function, the prefetch gets a page file name, the memory manager tries to validate the page file name and it is invalid and the system bugchecks.

-it would be nice to know the command line of the service to see what .exe is being spawned.
- find out if the superfetch or prefetch should even be on.
would have to look at each drive on your system, and maybe each device plugged into your USB ports.
 
Last edited:
note I added a note to my last post about turning off superfetch and prefetch
I just want to make sure you give it a try. Also, try to provide the kernel dump and I can see what actual service was calling these functions.
provide description of storage devices on the machine.

verify that TRIM is turned on: run cmd.exe as an admin then run
fsutil.exe behavior query DisableDeleteNotify

it should return
DisableDeleteNotify=0
(looking for a reason for the failed call)
 
Last edited:
last bugcheck was a watchdog timeout. the system was trying to read something from a pagefile and either the data was invalid or could not be loaded for some reason.

for these errors you would have to provide a kernel dump or even a full memory dump if the problem is malware or in the 32bit sub system.

note some of the compressed zip files had empty .dmp files in them.
 
Jul 27, 2022
10
0
10
0
last bugcheck was a watchdog timeout. the system was trying to read something from a pagefile and either the data was invalid or could not be loaded for some reason.

for these errors you would have to provide a kernel dump or even a full memory dump if the problem is malware or in the 32bit sub system.

note some of the compressed zip files had empty .dmp files in them.
thats odd, thought i was doing a kernel dump this whole time, my apologies. Hope i did this one right

https://1drv.ms/u/s!AvmvHsiIqhfdihDh_AzMFAn8y3AY?e=9Vxpaj

Ive also turned off prefetch and superfetch but i still got a bsod unfortunately. Made sure trim was on too

My storage consists of
Model ST1000DM003-1ER162 hdd
Model ADATA SX8200PNP
Dunno if my storage needs a firmware update or where to start for that

I also only have the essentials into my usb ports, like a keyboard and mouse

https://1drv.ms/u/s!AvmvHsiIqhfdihGqXnmL0V06kxEL?e=qwrDMl

Update: i think im gnna upgrade my cpu to a ryzen 5 5600g, best buy got a good deal, saving 110 bucks and its cheaper than my original cpu and is supposed to be the upgraded version of it haha
 
Last edited:
thats odd, thought i was doing a kernel dump this whole time, my apologies. Hope i did this one right

https://1drv.ms/u/s!AvmvHsiIqhfdihDh_AzMFAn8y3AY?e=9Vxpaj

Ive also turned off prefetch and superfetch but i still got a bsod unfortunately. Made sure trim was on too

My storage consists of
Model ST1000DM003-1ER162 hdd
Model ADATA SX8200PNP
Dunno if my storage needs a firmware update or where to start for that

I also only have the essentials into my usb ports, like a keyboard and mouse

https://1drv.ms/u/s!AvmvHsiIqhfdihGqXnmL0V06kxEL?e=qwrDMl

Update: i think im gnna upgrade my cpu to a ryzen 5 5600g, best buy got a good deal, saving 110 bucks and its cheaper than my original cpu and is supposed to be the upgraded version of it haha
buffer overflow in the kernel, relating to virtual memory.
chrome was running at the time. looks like it came out of the 32bit subsystem.
steam was updating a bunch of files.
one file hidden file you might look at:
\Users\qwerty\AppData\Local\Packages\Microsoft.XboxGamingOverlay_8wekyb3d8bbwe\LocalState\DiagOutputDir\XboxGamingOverlayTraces_FT_Server_20220731032756.txt

some discord module still doing something. only interesting if discord was uninstalled. in hidden app directory loaded from a roaming profile:
[\Users\qwerty\AppData\Roaming\discord\module_data\discord_voice\discord-webrtc_0]
here is a log file for the discord. you might take a look:
[\Users\qwerty\AppData\Roaming\discord\shared_proto_db\000003.log]

I guess at this point I would go and retest the cpu outside of windows.
maybe boot and run prime95 and see if the cpu runs ok.

this program
nvsphelper64.exe
is holding a lock that some other program is waiting to use the resource. (waited 8.9 million times)

i guess it is part nvidia share . disable it to see if it has an effect.
looks like part of nvidia shadowplay.

looks like chrome was holding the lock, same process that the memory manager called a bugcheck for.
(maybe some bad chrome extension?)

device for scarlet crush still connected but not loading due to registry error.
Dumping IopRootDeviceNode (= 0xffffe40ebc682010)
DevNode 0xffffe40ebc669bd0 for PDO 0xffffe40ebc683d40
InstancePath is "ROOT\SYSTEM\0002"
ServiceName is "ScpVBus"
State = DeviceNodeInitialized (0x302)
Previous State = DeviceNodeUninitialized (0x301)
Problem = CM_PROB_REGISTRY
Problem Status = 0xc0000034


I will look at the dump some more when I get back.
 
Last edited:
Jul 27, 2022
10
0
10
0
Thank you a lot

Found this under the xbox file
https://1drv.ms/t/s!AvmvHsiIqhfdihMYqtkrbwF_zgQL?e=dHJePj

discord file
https://1drv.ms/u/s!AvmvHsiIqhfdihIYjLlPIa7dWgzz?e=Lf00ud


hmm i do have some old chrome extensions, but the only ones enabled are adblock and octoshop (extension that price matches an item)

one thing to mention is that i did crash while on a discord call , so ive uninstalled discord again, should i delete most of the files that are still around even if discord is already uninstalled?

Update: Right when i tried to run a test on prime95, got another bsod
https://1drv.ms/u/s!AvmvHsiIqhfdihTuFjHBTKm8F2xC?e=ZSfMfy

another dmp https://1drv.ms/u/s!AvmvHsiIqhfdihVWmdRQlKu76BvS?e=XTc6Ue

https://1drv.ms/u/s!AvmvHsiIqhfdihaXmR278C7zJG3v?e=8AhbpL
 
Last edited:
Thank you a lot

Found this under the xbox file
https://1drv.ms/t/s!AvmvHsiIqhfdihMYqtkrbwF_zgQL?e=dHJePj

discord file
https://1drv.ms/u/s!AvmvHsiIqhfdihIYjLlPIa7dWgzz?e=Lf00ud


hmm i do have some old chrome extensions, but the only ones enabled are adblock and octoshop (extension that price matches an item)

one thing to mention is that i did crash while on a discord call , so ive uninstalled discord again, should i delete most of the files that are still around even if discord is already uninstalled?

Update: Right when i tried to run a test on prime95, got another bsod
https://1drv.ms/u/s!AvmvHsiIqhfdihTuFjHBTKm8F2xC?e=ZSfMfy

another dmp https://1drv.ms/u/s!AvmvHsiIqhfdihVWmdRQlKu76BvS?e=XTc6Ue

https://1drv.ms/u/s!AvmvHsiIqhfdihaXmR278C7zJG3v?e=8AhbpL
first minidump was chrome running but a bunch of user mode code then a call into the memory management to delete some page table entry. one of the windows core files: win32kfull.sys has been modified and checksum removed.
bugcheck is from the memory manager and is a undocumented code
0x402
errors in the 0x400 range relate to problems in page table.
for example 0x403 could be bit corruptions in cpu page table entries
Bug Check 0x1A MEMORY_MANAGEMENT - Windows drivers | Microsoft Docs

second bugcheck (0lder one)
shows chrome running then a call to memory manager to get a page.
the call gets a access violation and the system bugchecks.

note: win32full.sys has not been modified in this dump.

might be some chrome extension malware ?
or some cpu issue with pagefile access.

the xbox txt file only showed a bad parameter call for the game bar. looks like the handle was null so it the error would be expected.

log for discord was not in a text format, could not read it.

last bugcheck was a bugcheck in windows system process trying to modify a standby list page table control. standby memory is memory that windows preloads software into before you actually ask to run a file. you can use rammap64 to clear this list. I always figured this would be a great place to put malware. Anyway, the system process was making a call to these routines. deleting something from the list and got a bad memory address and bugchecked.

in this bugcheck 3 windows core files have been modified.
win32base.sys, wind32kfull.sys and win32k.sys


you might remove the game driver:
C:\Program Files\Riot Vanguard\vgk.sys Mon Jul 18 14:52:59 2022
using autoruns64.exe
just to be sure it is not doing something unexpected.
 
Last edited:
Jul 27, 2022
10
0
10
0
first minidump was chrome running but a bunch of user mode code then a call into the memory management to delete some page table entry. one of the windows core files: win32kfull.sys has been modified and checksum removed.
bugcheck is from the memory manager and is a undocumented code
0x402
errors in the 0x400 range relate to problems in page table.
for example 0x403 could be bit corruptions in cpu page table entries
Bug Check 0x1A MEMORY_MANAGEMENT - Windows drivers | Microsoft Docs

second bugcheck (0lder one)
shows chrome running then a call to memory manager to get a page.
the call gets a access violation and the system bugchecks.

note: win32full.sys has not been modified in this dump.

might be some chrome extension malware ?
or some cpu issue with pagefile access.

the xbox txt file only showed a bad parameter call for the game bar. looks like the handle was null so it the error would be expected.

log for discord was not in a text format, could not read it.

last bugcheck was a bugcheck in windows system process trying to modify a standby list page table control. standby memory is memory that windows preloads software into before you actually ask to run a file. you can use rammap64 to clear this list. I always figured this would be a great place to put malware. Anyway, the system process was making a call to these routines. deleting something from the list and got a bad memory address and bugchecked.

in this bugcheck 3 windows core files have been modified.
win32base.sys, wind32kfull.sys and win32k.sys


you might remove the game driver:
C:\Program Files\Riot Vanguard\vgk.sys Mon Jul 18 14:52:59 2022
using autoruns64.exe
just to be sure it is not doing something unexpected.
hi quick update, i upgraded my cpu to a ryzen 5 5600g and so far my computer has been running longer than before w/o a crash, so i think the cpu rlly did have a chipset bug or smth, thanks for the help for now, hopefully amd puts out an update or smth!
 

ASK THE COMMUNITY