Hi,
The latest attached crash dump from August 10th, is a 119: VIDEO_SCHEDULER_INTERNAL_ERROR bugcheck faulting dxgmms1.sys. To just give you an idea and understanding of what's going on, I will explain why DirectX is the fault and not the drivers.. but first, let's take a look at the stack trace:
Code:
STACK_TEXT:
fffff880`035ae758 fffff880`0455222f : 00000000`00000119 00000000`07000000 fffffa80`0aca2010 fffffa80`0e620af0 : nt!KeBugCheckEx
fffff880`035ae760 fffff880`03143318 : fffffa80`0e60ebb0 00000000`00000000 00000000`00000000 fffffa80`0e620b50 : watchdog!WdLogEvent5+0x11b
fffff880`035ae7b0 fffff880`0314337a : fffffa80`0000378e fffffa80`0e620b50 00000000`00000000 fffffa80`0aca2010 : dxgmms1!VidSchiCheckConditionDeviceCommand+0x8c
fffff880`035ae7e0 fffff880`0313f360 : fffffa80`06e10df0 fffffa80`0e620af0 fffffa80`0acc1000 fffffa80`0acc1000 : dxgmms1!VidSchiCheckPendingDeviceCommand+0x3e
fffff880`035ae820 fffff880`0313e7da : fffffa80`00000001 fffffa80`00000000 00000000`00000002 00000000`00000000 : dxgmms1!VidSchiProcessCompletedQueuePacketInternal+0x438
fffff880`035ae8e0 fffff880`0313de00 : fffffa80`00000000 fffffa80`0aca2010 00000000`00000000 fffff880`060c7000 : dxgmms1!VidSchiProcessDpcCompletedPacket+0x3b6
fffff880`035ae980 fffff880`0313dc4c : 00000000`00000002 fffffa80`0aca2010 00000000`00000000 00000000`00000000 : dxgmms1!VidSchDdiNotifyDpcWorker+0x198
fffff880`035ae9d0 fffff880`030441cf : fffffa80`09b53040 00000000`00000000 00000000`00000022 00000000`00000000 : dxgmms1!VidSchDdiNotifyDpc+0x94
fffff880`035aea20 fffff880`0f4e77f0 : fffffa80`09d76000 fffffa80`09d76000 00000000`00000000 00000000`00000000 : [SIZE=3][COLOR=Red][U][B]dxgkrnl!DxgNotifyDpcCB[/B][/U][/COLOR][/SIZE]+0x77
fffff880`035aea50 fffffa80`09d76000 : fffffa80`09d76000 00000000`00000000 00000000`00000000 fffff880`0f4e7783 : nvlddmkm+0xd97f0
fffff880`035aea58 fffffa80`09d76000 : 00000000`00000000 00000000`00000000 fffff880`0f4e7783 fffffa80`09d76000 : 0xfffffa80`09d76000
fffff880`035aea60 00000000`00000000 : 00000000`00000000 fffff880`0f4e7783 fffffa80`09d76000 00000000`00000000 : 0xfffffa80`09d76000
As you can see, I have bolded, underlined and italicized the important thing to look at here, which is dxgkrnl!DxgNotifyDpcCB.
Interrupts / Interrupt Service Routines to handle device I/O need to be done quickly or they run the risk of holding up the system due to the high IRQL. A simple and quick definition of IRQL is: An interrupt request level (IRQL) defines the hardware priority at which a processor operates at any given time.
If you'd like to further read into it, you can check this article:
What is IRQL? - A Hole In My Head - Site Home - MSDN Blogs
Due to the system running the risk of being held up, the interrupt is designed to create what is called a DPC (Deferred Procedure Call), which puts off the responsibility of handling the I/O till later. Just for a short reference, we can see Dpc in the stack trace as mentioned earlier: (dxgkrnl!DxgNotify
DpcCB).
Now, once the DPC itself is in the next DPC queue, it will then do the servicing of the I/O device. The interrupt is there to notify the system to prepare for the I/O, but the DPC is doing all the work.
Essentially, the interrupt has already done all of its work, and now the DPC must do its work (the I/O), which DirectX is involved in (and that is why you see DirectX as the fault in the crash dump).
So, with that said... we have a few things to do and try here:
1. Check nVidia's website and ensure you're on the latest drivers -
Drivers - Download NVIDIA Drivers
If you are already on the latest drivers, I recommend rolling back a version or so to be sure the latest drivers are not the issue. I do not recommend beta drivers unless they are a last resort in troubleshooting situations.
If rolling back a version or so does not help, given your previous crash dumps being PCI-e x124, your issue is likely most definitely motherboard related.
Regards,
Patrick