Re: [PATCH] vfio: Allow untainted use of driver_override

From: Yao Yuan

Date: Fri Nov 20 2026 - 19:42:51 EST


On Fri, Nov 20, 2026 at 03:10:38PM +0100, Uwe Kleine-König wrote:
> On Fri, Nov 20, 2026 at 10:17:46AM +0800, Yao Yuan wrote:
> > On Wed, Nov 18, 2026 at 04:56:44PM +0800, Alex Williamson wrote:
> > > Driver overrides are the standard mechanism for binding vfio drivers
> > > to devices. With the introduction of the support_driver_override flag
> > > in the core driver struct, it's now required for drivers to declare
> > > themselves as driver_override compatible, otherwise driver_override
> > > bindings will be denied or generate a kernel taint.
> > >
> >
> > Hi Alex,
> >
> > > Set the support_driver_override flag on the driver struct across all
> > > vfio drivers.
> >
> > LGTM.
> >
> > Reviewed-by: Yuan Yao <yaoyuan@xxxxxxxxxxxxxxxxx>
>

Hi Uwe,

> I missed your reply when I sent the next revision of this patch. I had

Thanks for your reminder, no big deal to me.
Let me follow your link to have a look on v2 there :-)

> to trim the recipients of the v2 because gmail refused to send it with
> everyone on Cc: who was emitted by get_maintainer.pl. So please see
> https://lore.kernel.org/kvm/fb0df5c89336b77a373ae0f890f63427e4adbb28.1795171088.git.u.kleine-koenig@xxxxxxxxxxxx
>
> But also note, that if v2 is picked up using `b4 am`, your Reviewed-by:
> is considered as the patch didn't change.


>
> Best regards
> Uwe