Re: [PATCH v5 03/17] iommu/arm-smmu-v3: Move arm_smmu_rmr_install_bypass_ste()

[Jason Gunthorpe <jgg@nvidia.com>]

On Thu, Feb 15, 2024 at 07:01:59PM +0000, Robin Murphy wrote:
> On 13/02/2024 3:37 pm, Mostafa Saleh wrote:
> > Hi Jason,
> > 
> > On Tue, Feb 06, 2024 at 11:12:40AM -0400, Jason Gunthorpe wrote:
> > > Logically arm_smmu_init_strtab() is the function that allocates and
> > > populates the stream table with the initial value of the STEs. After this
> > > function returns the stream table should be fully ready.
> > > 
> > > arm_smmu_rmr_install_bypass_ste() adjusts the initial stream table to force
> > > any SIDs that the FW says have IOMMU_RESV_DIRECT to use bypass. This
> > > ensures there is no disruption to the identity mapping during boot.
> > > 
> > > Put arm_smmu_rmr_install_bypass_ste() into arm_smmu_init_strtab(), it
> > > already executes immediately after arm_smmu_init_strtab().
> > > 
> > > No functional change intended.
> > 
> > I think arm_smmu_init_strtab is quite low level to abstract FW configuration in it.
> > For example in KVM[1] we'd re-use a big part of this driver and rely on similar
> > low-level functions. But no strong opinion.
> 
> Right, the fact that RMR handling is currently based on bypass STEs is an
> implementation detail; if we ever get round to doing the strict version with
> full-on temporary pagetables, that would obviously not belong in
> init_strtab, thus I would prefer to leave the "handle RMRs" step in its
> appropriate place in the higher-level flow regardless of how it happens to
> be named and implemented today.

I will drop this patch

Jason