Re: RFC: (re-)binding the VFIO platform driver to a platform device

[Scott Wood <scottwood@freescale.com>]

On Wed, 2013-10-02 at 21:13 +0100, Christoffer Dall wrote:
> On Wed, Oct 02, 2013 at 03:04:15PM -0500, Kim Phillips wrote:
> > On Wed, 2 Oct 2013 11:43:30 -0700
> > Christoffer Dall <christoffer.dall@linaro.org> wrote:
> > 
> > > On Wed, Oct 02, 2013 at 01:32:38PM -0500, Scott Wood wrote:
> > > > What's wrong with a non-vfio-specific flag that a driver can set, that
> > > > indicates that the driver is willing to try to bind to any device on the
> > > > bus if explicitly requested via the existing sysfs bind mechanism?
> > > > 
> > > It sounds more hackish to me to invent some 'generic' flag to solve a
> > > very specific case.  What you're suggesting would let users specify that
> > > a serial driver should handle a NIC hardware, no?  That sounds much much
> > > worse to me.
> > 
> > I thought that was the nature of VFIO drivers...it's a 'meta-' driver,
> > used for enabling userspace drivers at large.
> > 
> Yes, vfio is a meta driver, therefore it needs to be able to do
> something special, but the generic driver/device/bus matching framework
> doesn't need an extra generic feature allowing you to bind driver X to
> device Y for all combinations of X and Y depending on  some flag...

Not all combinations of X and Y.  Only instances of X that advertise
that this is OK.

> Someone please correct me if there are more use cases for this and this
> is in fact worth a generic solution.

Note that the wildcard match that I suggested in the e-mail I just sent
would likely be implemented by the bus match code -- not by generic
driver model code.  It would still be less intrusive than implementing a
dynamic match mechanism for each bus type (and for device tree, ACPI,
etc in the case of platform bus).

-Scott