Re: [PATCH v4 0/6] iommufd: Iterate the cache invalidation array in the core

[Will Deacon]

On Thu, Jul 16, 2026 at 09:20:50AM -0700, Nicolin Chen wrote:
> On Thu, Jul 16, 2026 at 12:18:45PM -0300, Jason Gunthorpe wrote:
> > On Wed, Jul 15, 2026 at 01:30:22PM -0700, Nicolin Chen wrote:
> > > On Wed, Jul 15, 2026 at 05:15:21PM -0300, Jason Gunthorpe wrote:
> > > > On Tue, Jul 14, 2026 at 11:48:46AM -0700, Nicolin Chen wrote:
> > > > > Nicolin Chen (6):
> > > > >   iommu/arm-smmu-v3: Support IDR5.DS and widen the TLBI SCALE field
> > > > >   iommu/arm-smmu-v3-iommufd: Reject unsupported bits in invalidation
> > > > >     commands
> > > > >   iommufd: Iterate the cache invalidation array in the core
> > > > >   iommufd/selftest: Convert cache invalidation mocks to the core array
> > > > >     loop
> > > > >   iommu/arm-smmu-v3-iommufd: Convert cache invalidation to the core
> > > > >     array loop
> > > > >   iommu/vt-d: Convert nested cache invalidation to the core array loop
> > > > 
> > > > I was tempted to take this through the iommufd tree, but Will, I think
> > > > the first two ought to go to -stable through iommu ?
> > > > 
> > > > If you do that then maybe I'll pick up the rest
> > > 
> > > PATCH-5 touches arm-smmu-v3-iommufd. Merging via the iommufd tree
> > > will conflict with the PATCH-2 in Ashish's series (via iommu):
> > > https://lore.kernel.org/all/20260714104202.1664187-1-amhetre@xxxxxxxxxx/
> > > 
> > > Maybe it needs a shared branch; and then, Ashish can rebase?
> > 
> > Urk this is a pain. I think Ashish's errata fix should go first so it
> > is easier to backport.
> 
> OK. I can rebase and send a v5 after Will takes Ashish's series.

I've now queued the workaround from Ashish.

Will