Re: [PATCH v2] driver core/ACPI: Introduce companion_bus_register()

[Rafael J. Wysocki (Intel)]

On Wed, Sep 23, 2026 at 8:50 PM Rafael J. Wysocki (Intel)
<rafael@xxxxxxxxxx> wrote:
>
> On Wed, Sep 23, 2026 at 8:31 PM Rafael J. Wysocki (Intel)
> <rafael@xxxxxxxxxx> wrote:
> >
> > On Wed, Sep 23, 2026 at 7:32 PM Rafael J. Wysocki (Intel)
> > <rafael@xxxxxxxxxx> wrote:
> > >
> > > On Wed, Sep 23, 2026 at 7:06 PM Michael Kelley <mhklinux@xxxxxxxxxxx> wrote:
> > > >
> > > > From: Rafael J. Wysocki <rafael@xxxxxxxxxx> Sent: Wednesday, September 9, 2026 9:41 AM
> > > > >
> > > > > The ACPI bus type does not allow drivers to be registered, so the sysfs
> > > > > attributes related to drivers created for it and its devices are
> > > > > useless, and its drivers/ directory is always empty.  All of that is
> > > > > confusing and wasteful.
> > > > >
> > > > > To allow skipping the creation of those sysfs attributes, introduce a
> > > > > "companion" bus type concept and add a special registration function for
> > > > > registering "companion" bus types, companion_bus_register().
> > > > >
> > > > > Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@xxxxxxxxx>
> > > >
> > > > As a heads up, this patch breaks Linux guests booting on Hyper-V.
> > > > Something goes wrong in the VMBus driver getting loaded and
> > > > initialized (drivers/hv/vmbus_drv.c).
> > >
> > > This patch doesn't affect anything except for bus registration and
> > > driver addition and lookup for the ACPI bus type.
> > >
> > > > It may be that the VMBus
> > > > driver is doing something wrong or making an invalid assumption.
> > > > I'll debug the problem today or tomorrow, but any insight you can
> > > > offer would be appreciated. I'm working with linux-next20260921.
> > >
> > > No idea what's going on.
> >
> > Actually, you may try to restore the stub .match() callback for
> > acpi_bus_type because the lack of it causes 1 to be returned by
> > driver_match_device(), so if any driver points to acpi_bus_type as its
> > bus by mistake, the check will now succeed.
>
> Which shouldn't matter though because driver_match_device() is only
> called in 3 places and none of them is reachable for acpi_bus_type
> after the $subject patch AFAICS.

Well, not quite.

If a driver with an incorrect bus is passed to driver_attach(), it
will iterate over devices on that bus and call __driver_attach() for
each of them.  If driver_match_device() succeeds then, the device will
be passed to driver_probe_device() and it will not be good.

So it may be worth checking after all.