Windows 10 BSOD - KERNEL_MODE_HEAP_CORRUPTION

[Shonk1]

The last few months i have been getting random bsod's when watching youtube Kernel Mode Heap Corruption
and have been pulling my hair out trying to find the issue

e.g. i fall asleep watching youtube and wake up to it rebooted

To cut a long story short i installed a fresh install of Windows 10 20H2 on a secondary drive
and ran all inbox drivers

Ran all my stability tests

(A reliable ish way of triggering this in an hour or two)
Have firefox playing youtube in the background whilst running MemTestPro 7.0 16 Threads x 3800MB (I have 64GB Ram)

It was 100% fine with inbox drivers

Installed NVME drivers (Samsung)
Installed Installed Intel RST Drivers

Ran all my stability tests

Within 1 hour
BSOD Kernel Mode Heap Corruption

Removed Samsung NVME Drivers and ran inbox Microsoft NVME Drivers
Rebooted

Ran all my stability tests

Within 1 hour
BSOD Kernel Mode Heap Corruption

Removed Intel RST Drivers and used Inbox RST Drivers
Ran all my stability tests

got to 500% ram test (a long time maybe 7 hours) it was fine
rebooted
ran again
left it to 850% (12 hours or so) also fine

I have used various RST drivers
17.9.0.1007
17.9.1.1009
17.9.2.1013
17.9.4.1017

all do it in Raid mode

I have switched to Ahci mode for now with 17.9.0.1007

as the sata drives are just empty anyway atm

and it seems fine

I am listing a few things that may help

I dont have the Intel RST UWP app installed
RstMwService is set to manual in services(RPC endpoint service which allows communication between driver and Windows Store Application)

Telemetry is disabled
; ********** Startup Trace Session Intel RST Delete **********
[HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\WMI\Autologger\IntelRST]
"Start"=dword:00000000
"Status"=dword:00000000

[HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\StorPort]
"TelemetryPerformanceEnabled"=dword:00000000
"TelemetryErrorDataEnabled"=dword:00000000
"TelemetryDeviceHealthEnabled"=dword:00000000

CPU = Intel Core i9 9900KS
Motherboard 1 = Gigabyte Z390 Aorus Master Bios F11L
Motherboard 2 = Gigabyte Z390 M Gaming Bios F9L (i replaced this with above trying to fix the issue)
Ram = 4 x 16GB Crucial Ballistix Sport LT 3200 (Yes it does it with the ram at 2400/2666)
M2M = Sabrent Rocket TLC 1TB NVME SSD (OS Drive)
M2A = Sabrent Rocket TLC 1TB NVME SSD
M2P = Sabrent Rocket TLC 2TB NVME SSD
PCIEX1_1 = Intel Optane 800P 118GB NVME SSD (Scratch Drive) (not in optane mode and not using intel drivers)
PCIEX1_2 = Corsair MP510 480GB NVME SSD
PCIEX1_3 = Toshiba XG3 512GB NVME SSD
SATA0 = Adata SU800 512GB (Raid0 with other SU800)
SATA1 = Adata SU800 512GB
SATA2 = Samsung 860 Evo 500GB (Raid0 with other 860 Evo)
SATA3 = Samsung 860 Evo 500GB

Someone else has been getting this problem on a Z490

https://community.intel.com/t5/Rapid-Storage-Technology/Windows-10-BSOD-KERNEL-MODE-HEAP-CORRUPTION-iaStorAC/m-p/1201031#M8613

https://community.intel.com/t5/Rapid-Storage-Technology/Windows-10-BSOD-KERNEL-MODE-HEAP-CORRUPTION-iaStorAC-again/td-p/1210167

This needs fixing please pass this onto the dev's

BTW MY old Intel DQ77MK Q77 with Xeon E3 1290 V2 runs just fine with 17.9.xxx drivers

[Esteban_D_Intel]

Hello Shonk1,

Thank you for posting on the Intel® communities. 

First, I would like to thank you for the complete information provided.

To better assist you, I would appreciate if you could provide the following:

• What changes were made prior to this behavior.
• For how long you’ve been experiencing this issue?

Please run the following tool and attach the reports generated.   

Intel® System Support Utility (Intel® SSU) Download link 

1.Open the application and click on "Scan" to see the system and device information. 

2.By default, Intel® SSU will take you to the "Summary View".   

3.Click on the menu where it says "Summary" to change to "Detailed View".   

4.To save your scan, click on "Next", then "Save".  

Please follow the System Reports for Intel® Rapid Storage Technology to gather the RST report.

I would like to further investigate this behavior for you, please reply at your earliest convinience with the information attached.

Esteban D.

Intel Technical Support Technician 

[Shonk1]

Hi thanks for the quick reply

It seems the pc isnt quite as stable as i though

Gigabyte set the CPU Internal AC/DC loadline on my Aorus Z390 Master to Auto which is 1.300 / 1.300 mOhm apparently

Whilst it passes a full load stability test
it seems when the cpu is bouncing around its not quite 100% but very close

I think maybe my specific i9-9900KS is a bit to close to the redline / a poor bin/sample

and gigabyte didnt fully test Auto with a wide range of the i9-9900KS

The intel Spec for IA Domain Loadline (AC/DC) is 1.600 / 1.600 mOhm

I have set my AC/DC Loadline to intel spec 1.600 / 1.600 mOhm now

My CPU Vcore Loadline Calibration was already at intel spec 1.6 mOhms

Maybe RST is very sensitive to this where as the inbox RST where maybe not so much

I have set the Sata Drive back up as raid and installed RST 17.9.4.1017

Give me a day or two to see if it still does it

[Shonk1]

If it does end up being this it doesnt say much for intel's "Thin Bin" if it isnt fully stable with CPU Internal AC/DC loadline 1.300 / 1.300 mOhm all board makers default to this on Z390 and im sure Z490 also

[Esteban_D_Intel]

Hello Shonk1,

Thank you so much for your response.

I understand that you require an additional day or two to test your current configuration and check if the behavior persists. 

Feel free to provide an update to this thread once you have it and we’ll be more than glad to continue with the assistance.

Also, we would appreciate if you could attach the reports previously requested.

Esteban D.

Intel Technical Support Technician  

[Esteban_D_Intel]

 Hello Shonk1

Were you able to check the previous post?   

Please let me know if you need further assistance.   

Esteban D.  

Intel Technical Support Technician   

[Shonk1 · Accepted solution]

I was testing

Intel RST 18.30.0.1135

which seemed fine then i noticed

Intel RST 18.30.1.1138

had been released which also seems fine

[Esteban_D_Intel]

Hello Shonk1,

Thank you so much for your response.

I’m glad to know that you were able to resolve this behavior that way. Hopefully, this resolution will be helpful to other community members.

If you need any additional information, please submit a new question as this thread will no longer being monitored.

Esteban D.  

Intel Technical Support Technician