[PATCH 0/2] Improve vfio-pci primary GPU assignment behavior

[Gerd Hoffmann]

  Hi,

> >> But also, this issue isn't something that only affects graphic devices,
> >> right? AFAIU from [1] and [2], the same issue happens if a PCI device
> >> has to be bound to vfio-pci but already was bound to a host driver.
> > 
> > Nope.  There is a standard procedure to bind and unbind pci drivers via
> > sysfs, using /sys/bus/pci/drivers/$name/{bind,unbind}.
> >
> 
> Yes, but the cover letter says:
> 
> "Users often employ kernel command line arguments to disable conflicting
> drivers or perform unbinding in userspace to avoid this"

Thats helpful at times to deal with driver and/or hardware quirks.
Example: Years ago drm drivers used to be horrible when it came to
unbind, leaving oopses and panics left & right when you tried (luckily
it works much better these days).

[ leaving this here for completeness, snipping the remaining reply,
  noting that we are on the same page now ]

thanks & take care,
  Gerd