Re: [PATCH v11 05/11] iommu/arm-smmu-v3: Disable the EVTQ and the PRIQ in a kdump kernel

[Nicolin Chen <nicolinc@nvidia.com>]

On Tue, Oct 20, 2026 at 11:22:46AM +0000, Mostafa Saleh wrote:
> On Mon, Oct 05, 2026 at 12:30:48PM -0700, Nicolin Chen wrote:
> > @@ -5576,6 +5573,15 @@ static int arm_smmu_device_hw_probe(struct arm_smmu_device *smmu)
> >  	if (arm_smmu_sva_supported(smmu))
> >  		smmu->features |= ARM_SMMU_FEAT_SVA;
> >  
> > +	/*
> > +	 * A kdump kernel wants neither queue: the crashed kernel's CDs and page
> > +	 * tables might be corrupted, spamming events, and page requests cannot
> > +	 * be served. A disabled queue discards new records without raising any
> > +	 * global error.
> > +	 */
> 
> I find this a bit confusing. If the CD or page table of the crashed kernel are
> corrupted how is it safe to adopt them later in this series?

The adoption code runs some sanity test against STE and CD. But we
have no guarantee that every bit in CD and PTEs is sane.

The point here is that, if one bit in CD or PTEs is ever corrupted,
at least we don't spam the system with events.

Nicolin