Re: [PATCH 01/11] iommu: Add device dma ownership set/release interfaces

[Lu Baolu]

Hi Joerg,

On 11/19/21 11:06 PM, Jörg Rödel wrote:
> On Fri, Nov 19, 2021 at 07:14:10PM +0800, Lu Baolu wrote:
> > The singleton group requirement for iommu_attach/detach_device() was
> > added by below commit:
> >
> > commit 426a273834eae65abcfc7132a21a85b3151e0bce
> > Author: Joerg Roedel <jroedel@xxxxxxx>
> > Date:   Thu May 28 18:41:30 2015 +0200
> >
> >      iommu: Limit iommu_attach/detach_device to devices with their own group
> >
> >      This patch changes the behavior of the iommu_attach_device
> >      and iommu_detach_device functions. With this change these
> >      functions only work on devices that have their own group.
> >      For all other devices the iommu_group_attach/detach
> >      functions must be used.
> >
> >      Signed-off-by: Joerg Roedel <jroedel@xxxxxxx>
> >
> > Joerg,can you please shed some light on the background of this
> > requirement? Does above idea of transition from singleton group
> > to group with single driver bound make sense to you?
>
> This change came to be because the iommu_attach/detach_device()
> interface doesn't fit well into a world with iommu-groups. Devices
> within a group are by definition not isolated between each other, so
> they must all be in the same address space (== iommu_domain). So it
> doesn't make sense to allow attaching a single device within a group to
> a different iommu_domain.

Thanks for the explanation. It's very helpful. There seems to be a lot
of discussions around this, but I didn't see any meaningful reasons to
break the assumption of "all devices in a group being in a same address
space".

Best regards,
baolu

> I know that in theory it is safe to allow devices within a group to be
> in different domains because there iommu-groups catch multiple
> non-isolation cases:
>
> 	1) Devices behind a non-ACS capable bridge or multiple functions
> 	   of a PCI device. Here it is safe to put the devices into
> 	   different iommu-domains as long as all affected devices are
> 	   controlled by the same owner.
>
> 	2) Devices which share a single request-id and can't be
> 	   differentiated by the IOMMU hardware. These always need to be
> 	   in the same iommu_domain.
>
> To lift the single-domain-per-group requirement the iommu core code
> needs to learn the difference between the two cases above.
>
> Regards,
>
> 	Joerg