Question Display driver amdkmdap keeps crashing

[Boutchard]

So I am posting here since It's been 3 days since I am working on this issue. I am having BSOD after BSOD after BSOD, here are the code of my bsod(s):

-SYSTEM_SERVICE_EXCEPTION
-MEMORY_MANAGEMENT
-KMODE_EXCEPTION_NOT_HANDLED

Here are my specs:

-CPU : AMD Ryzen 5 2600 6 cores
-RAM: 16 GB Corsair vengeance 2400 mHz (DISCLAIMER : these are brand new, i had one dead stick so I RMAed it to corsair and they sent me a new kit about a week ago)
-GPU : Sapphire Nitro+ Radeon Rx 580 4Gb
-Mobo : MSI B450I Gaming Plus AC mini ATX
-PSU : SeaSonic M12II 520 Bronze Evo Edition
-SSD(on wich the os is installed) : SAMSUNg SSD 850 evo 120 Gb
-Hdd (On wich i installed games to test) : Western Digital Blue 1 TB

Every part is less than 10 months old( invoice to prove it)from a very good and trustable friend, I immediatly noticed that a ram stick was dead. I than used the PC for a whole week on 1 ram stick without any kind of problem than i RMAed it. Ever since, i constantly get BSOD for multiple reason mainly while playing games. Here s is what i tried:

-Switching Graphics card
-Reinstalling OS
-Updated Bios
-Only using 1 RAM stick on either of the slots
-Reinstalled OS again
-Memcheck
-Chkdsk(Altough in the log files it is said to have found corrupt files)

Also when i run sfc /scannow , it says that it found corrupt files but that it couldnt fix them. Here are the module that caused the BSOD:

• ntoskrnl.exe
• ntfs.sys
• hardware.sys

I am going to be honest, i tried to copy the OS from my old pc to my new with a free online tool (before completely wiping and reinstalling) since the os on the new PC wouldnt even boot(this was after the 1 week prior to the RMA just before RMAing the RAM)


Any help would be immensely appreciated

 

[Colif]

1. Can you follow option one on the following link - here - and then do this step below: Small memory dumps - Have Windows Create a Small Memory Dump (Minidump) on BSOD

that creates a file in c windows/minidump after the next BSOD
copy that file to documents
upload the copy from documents to a file sharing web site, and share the link here and I will get someone to convert file into a format I can read

2. Have you tried memtest? Not just the memory checker in windows
Try running memtest86 on each of your ram sticks, one stick at a time, up to 4 passes. Only error count you want is 0, any higher could be cause of the BSOD. Remove/replace ram sticks with errors. It creates a bootable USB

3. Have you downloaded and run live update 6 from here - https://www.msi.com/Motherboard/support/B450I-GAMING-PLUS-AC#down-driver&Win10 64 (its on the utility tab)

 

[cherry blossoms]

As Colif indicated, the actual dmp file would be helpful to identify the bugcheck parameters

Also to check:
Run the Samsung magician software and check your drive health and firmware?

Samsung Magician Software Download | Samsung Semiconductor Global

Samsung Magician software is designed to help you manage your Samsung SSD with a simple, intuitive user interface. Download files & find supported models.

www.samsung.com

By wipe and reinstall, I am assuming you downloaded a fresh copy of windows 10 using the creation tool?
https://www.microsoft.com/en-us/software-download/windows10

Your other running programs not specified. A clean boot might also help you narrow down the problem if you have 3rd party software installed:
https://support.microsoft.com/en-us/help/929135/how-to-perform-a-clean-boot-in-windows

 

[Boutchard]

1. Can you follow option one on the following link - here - and then do this step below: Small memory dumps - Have Windows Create a Small Memory Dump (Minidump) on BSOD

that creates a file in c windows/minidump after the next BSOD
copy that file to documents
upload the copy from documents to a file sharing web site, and share the link here and I will get someone to convert file into a format I can read

2. Have you tried memtest? Not just the memory checker in windows
Try running memtest86 on each of your ram sticks, one stick at a time, up to 4 passes. Only error count you want is 0, any higher could be cause of the BSOD. Remove/replace ram sticks with errors. It creates a bootable USB

3. Have you downloaded and run live update 6 from here - https://www.msi.com/Motherboard/support/B450I-GAMING-PLUS-AC#down-driver&Win10 64 (its on the utility tab)

Just for info, I changed my SSd for a regualr HDD but still get crashes
1- Heres is the link : https://drive.google.com/drive/folders/1_UqYOaJqZh73-OQ47q3HPwySHXpgf5tZ?usp=sharing

2- No how long should i run it? I will try to run in tonight

3 - No I will do it too

As Colif indicated, the actual dmp file would be helpful to identify the bugcheck parameters

Also to check:
Run the Samsung magician software and check your drive health and firmware?

Samsung Magician Software Download | Samsung Semiconductor Global

Samsung Magician software is designed to help you manage your Samsung SSD with a simple, intuitive user interface. Download files & find supported models.

www.samsung.com

By wipe and reinstall, I am assuming you downloaded a fresh copy of windows 10 using the creation tool?
https://www.microsoft.com/en-us/software-download/windows10

Your other running programs not specified. A clean boot might also help you narrow down the problem if you have 3rd party software installed:
https://support.microsoft.com/en-us/help/929135/how-to-perform-a-clean-boot-in-windows

Its not the SSD the problem, I swapped my ssd for an HDD and still get the bug

Yes i downloaded a fresh copy of windows with the creation tool (via bootable usb)

 

[gardenman]

Hi, I ran the dump files through the debugger and got the following information: https://pste.eu/p/d4F8.html

Spoiler

File information:	073119-29718-01 - Copie.dmp (Jul 31 2019 - 06:47:27)
Bugcheck:	MEMORY_MANAGEMENT (1A)
Probably caused by:	memory_corruption (Process: backgroundTask)
Uptime:	0 Day(s), 4 Hour(s), 04 Min(s), and 17 Sec(s)

File information:	073119-37250-01 - Copie.dmp (Jul 31 2019 - 07:19:58)
Bugcheck:	KMODE_EXCEPTION_NOT_HANDLED (1E)
Probably caused by:	memory_corruption (Process: VSSVC.exe)
Uptime:	0 Day(s), 0 Hour(s), 31 Min(s), and 56 Sec(s)

BIOS info was not included in the dump file. This can sometimes mean an outdated BIOS is be used.

This information can be used by others to help you. I can't help you with this. Someone else will post with more information. Please wait for additional answers. Good luck.

 

[Boutchard]

Ran memtest86 on one of the ram stick and it gave me 188 error on test 9... I switched to the other RAM stick (in the same slot) and will run memtest86 with it too. If it's giving me error too, I think it safe to assume that the mobo is the cause of those crash(even more because the RAM stick are brand new)

 

[Colif]

It could be motherboard or you just really unlucky with ram - I wonder, can you test ram in another PC and confirm the stick with 188 errors is actually okay?

Max error count you want in memtest = 0. ANY higher is too high, so 188 is way out.

are you running in XMP (I assume AMD boards have xmp support)

Test 9 [Modulo 20, Random pattern]

Using the Modulo-X algorithm should uncover errors that are not detected by moving inversions due to cache and buffering interference with the algorithm.

https://www.memtest86.com/technical.htm

most cases on test 9 is bad ram.

 

[Boutchard]

XMP is disabled, I ran memetest86 on the other RAM stick in the exact same slot and it showed no error.. I am going to rerun memtest86 with both RAM stick in the other slot.. if the same one is giving me error again it is safe to say I got a faulty ram stick yet again...

 

[Boutchard]

So I finally found what the problem was (well I am 95% sure) I did memtest86 on both stick of ram, 1 stick at a time in each slot, one stick Showed 0 error on both test, the other one showed 188 errors and over 1000+ errors, I am almost sure that I have yet again a bad RAM stick. I am going to rma it once again but is there anyway to make sure that my MOBO isnt killing my RAM since its the second RAM stick that I have to send back?

 

[Colif]

I don't think motherboards can kill ram... I know PSU can but Seasonic generally don't make bad PSU
Ram can kill motherboards, but not sure about other way. Google search doesn't show any examples from last 10 years

I will ask others if they seen it happen

 

[Boutchard]

Well now strangely I am only running on the working RAM stick and from time to time, my display appears to disconnect for a few seconds.. D you think it might be because i installed stuff while running on the corrupted RAM stick?

 

[Colif]

I would be tempted to get a technician to look at it, at very least he might have spare ram he can put into PC that he knows works, to check if you still get issues. It could be caused by another component

A bad VRM or VRM component causing excess voltage to go to things, sure, I could see that killing things plugged in. Though on recent hardware the memory is pretty much handled by the CPU, so it would have to be CPU failure to really kill RAM.

A short or leakage or something on the motherboard itself to the memory pins maybe. Certainly seen a few misplaced screws ruin people's day.

 

[Boutchard]

Used it for a good day without bsod but still that weird bug where screen disconnect...

EDIT: https://drive.google.com/file/d/15sXJbME5USTIWgfoXwUfQvqpRhG6H6BB/view?usp=sharing these are my GPU metrics when the games are crashing... Are these drop because of the crash or are the crashes because of the drop? Also right after the game crashes, the Radeon settings app just wont open

 

[Colif]

look in reliability monitor, are there any red X recently?
click on one,
does the bottom panel look like this:

Click on view technical details link next to data
does it say something like
"Display driver failed to start; using Microsoft Basic Display Driver instead. Check Windows Update for a newer display driver."

as i am helping someone else with a RX 580 having similar problems.. Screenshot is mine, I had a similar problem last year but I have an Nvidia GPU. MIne would try to restart the GPU drivers 9 times and then just give up and use basic, and turn screen off (as basic display driver doesn't like my 4k screen), or in other guys example, Crash to desktop.

MIne wasn't a GPU problem at all. Was something else in PC setting it off. Would you believe a mouse was the cause?

 

[ryzengamer]

Test with Debian buster live to check hardware failures. The dmesg command shows the kernel log.
https://cdimage.debian.org/debian-c...rid/debian-live-10.0.0-amd64-xfce.iso.torrent

 

[Boutchard]

Test with Debian buster live to check hardware failures. The dmesg command shows the kernel log.
https://cdimage.debian.org/debian-c...rid/debian-live-10.0.0-amd64-xfce.iso.torrent

Hummm... Is it going to work with my Windows Os or do i have to install a Debian Distros?

look in reliability monitor, are there any red X recently?
click on one,
does the bottom panel look like this:

Click on view technical details link next to data
does it say something like
"Display driver failed to start; using Microsoft Basic Display Driver instead. Check Windows Update for a newer display driver."

as i am helping someone else with a RX 580 having similar problems.. Screenshot is mine, I had a similar problem last year but I have an Nvidia GPU. MIne would try to restart the GPU drivers 9 times and then just give up and use basic, and turn screen off (as basic display driver doesn't like my 4k screen), or in other guys example, Crash to desktop.

MIne wasn't a GPU problem at all. Was something else in PC setting it off. Would you believe a mouse was the cause?

I dont see that error message, only a BSOD event has been logged saying "Hardware Error and "Shut down unexpectedly"
https://drive.google.com/file/d/15sXJbME5USTIWgfoXwUfQvqpRhG6H6BB/view?usp=sharing
Tested it out,exactly when my game crashe, this error message pop up in the reliability monitor:

Description
A problem with your hardware caused Windows to stop working correctly.

Problem signature
Problem Event Name: LiveKernelEvent
Code: 141
Parameter 1: ffff9b888aec4010
Parameter 2: fffff80622510750
Parameter 3: 0
Parameter 4: 1470
OS version: 10_0_18362
Service Pack: 0_0
Product: 256_1
OS Version: 10.0.18362.2.0.0.256.121
Locale ID: 4105

Also, the windows logger log this event during the crash:

Fault bucket 1228708542630178398, type 5
Event Name: RADAR_PRE_LEAK_64
Response: Not available
Cab Id: 0

 

[ryzengamer]

Hummm... Is it going to work with my Windows Os or do i have to install a Debian Distros?

Linux Live installers allows you to boot and use your computer from a USB memory stick without doing anything to your main drive. You have a full desktop running and can install Unigine Heaven and Superposition benchmarks to stress test your computer. If you like to use a Linux distribution, you can install it to your main drive. The Grub menu allows you to multi boot to different operating systems. Advanced users do not need other OS than Debian. Windows games works very well with wine-staging and DXVK and there are thousands of native Linux games: Rocket League, Tomb Raider 2013 and 2016, Portal 2, The Talos Principle, etc.

 

[Boutchard]

Linux Live installers allows you to boot and use your computer from a USB memory stick without doing anything to your main drive. You have a full desktop running and can install Unigine Heaven and Superposition benchmarks to stress test your computer. If you like to use a Linux distribution, you can install it to your main drive. The Grub menu allows you to multi boot to different operating systems. Advanced users do not need other OS than Debian. Windows games works very well with wine-staging and DXVK and there are thousands of native Linux games: Rocket League, Tomb Raider 2013 and 2016, Portal 2, The Talos Principle, etc.

I think this wont change anything since I suspect an hardware component to be faulty...

 

[Colif]

RADAR_PRE_LEAK_64 ... i can't find a lot on this. seems it could be a memory leak possibly tied to directX. Memory leaks are normally driver errors, DirectX errors are normally the GPU drivers

Guess that makes sense since driver is crashing

Where did you get radeon drivers from? Are they newest from AMD?

 

[Boutchard]

RADAR_PRE_LEAK_64 ... i can't find a lot on this. seems it could be a memory leak possibly tied to directX. Memory leaks are normally driver errors, DirectX errors are normally the GPU drivers

Guess that makes sense since driver is crashing

Where did you get radeon drivers from? Are they newest from AMD?

On the official AMD website, https://www.amd.com/en/support , everytime I installed it, I also ran DDU in safe mode before.It crashes with either of my rx 580 and my old r7 370, I switched even the PSU, still get the same crash, the only component that I haven't been able to switch is basically my CPU and my MOBO...

 

[Colif]

its a shame Ryzen CPU don't have graphics cores.

If I had to guess which of those 2 - CPU & Motherboard - I would lean towards motherboard only as it would explain the ram thing and why 2 AMD GPU give you similar errors.

but I don't like guessing, so that technician idea comes back again.

you can create this on another PC and run it on the PC without windows to test CPU - https://www.infopackets.com/news/10113/how-fix-bootable-prime95-stress-test-hardware

no way to test Motherboard though. Its why you test everything else and once you know the other items work, it has to be whatever is left.

Running linux would tell use if it is hardware as if you get errors on 2 operating systems, it removes windows from equation.

 

[Boutchard]

I am going to install a distros and test it on it. Could it be possible that when I burned the windows installation image on the usb key, I had the corrupt RAM stick, therefore also corrupting the bottable usb key itself? maybe reinstalling the ISO on the usb key than reinstalling windows with it would help?

 

[Colif]

i would test with linux before blaming the iso you used to install windows. If LInux has no problems, then you can try another iso

 

[Boutchard]

Welp, tested it in linux, still gets the crash...

 

[Colif]

if there was a test for motherboard I would offer it.

You either guess its the motherboard or get a tech to look at it.

Unless its something really random, we tested everything. I would try a tech