> From: David Gibson <david@xxxxxxxxxxxxxxxxxxxxx>
> Sent: Thursday, October 14, 2021 1:24 PM
>
> On Sun, Sep 19, 2021 at 02:38:41PM +0800, Liu Yi L wrote:
> > From: Lu Baolu <baolu.lu@xxxxxxxxxxxxxxx>
> >
> > These two helpers could be used when 1) the iommu group is singleton,
> > or 2) the upper layer has put the iommu group into the secure state by
> > calling iommu_device_init_user_dma().
> >
> > As we want the iommufd design to be a device-centric model, we want to
> > remove any group knowledge in iommufd. Given that we already have
> > iommu_at[de]tach_device() interface, we could extend it for iommufd
> > simply by doing below:
> >
> > - first device in a group does group attach;
> > - last device in a group does group detach.
> >
> > as long as the group has been put into the secure context.
> >
> > The commit <426a273834eae> ("iommu: Limit
> iommu_attach/detach_device to
> > device with their own group") deliberately restricts the two interfaces
> > to single-device group. To avoid the conflict with existing usages, we
> > keep this policy and put the new extension only when the group has been
> > marked for user_dma.
>
> I still kind of hate this interface because it means an operation that
> appears to be explicitly on a single device has an implicit effect on
> other devices.
>
I still didn't get your concern why it's such a big deal. With this proposal
the group restriction will be 'explicitly' documented in the attach uAPI
comment and sample flow in iommufd.rst. A sane user should read all
those information to understand how this new sub-system works and
follow whatever constraints claimed there. In the end the user should
maintain the same group knowledge regardless of whether to use an
explicit group uAPI or a device uAPI which has group constraint...
Thanks,
Kevin
