Re: IRP_MN_SET_POWER - will do some more digging.

[Paul Durrant <xadimgnik@xxxxxxxxx>]

On 17/10/2022 14:59, Martin Harvey wrote:
> Okay, well regardless of discussion about the power model, we still see these 
> occasionally, it's an open question at the moment as to whether it's the IRP 
> handling (which you have nicely serialized), or whether pci is expecting some 
> different ordering somewhere.
>
> I can probably find time to do some more digging.
TBH that PciStallForPowerChange suggests it is waiting for something in 
config space. I'm guessing from nvlddmkm in the stack that was a vGPU 
VM? I'd be looking at the config space emulation for my problem, not the 
PV drivers :-)
  Cheers,

    Paul