[PATCH v2 05/45] KVM: arm64: GICv3: Detect and work around the lack of ICV_DIR_EL1 trapping

[Oliver Upton]

On Thu, Nov 13, 2025 at 08:10:18PM +0000, Marc Zyngier wrote:
> +Fuad
> 
> On Thu, 13 Nov 2025 19:06:31 +0000,
> Mark Brown <broonie at kernel.org> wrote:
> > 
> > [1  <text/plain; us-ascii (7bit)>]
> > On Thu, Nov 13, 2025 at 06:15:29PM +0000, Marc Zyngier wrote:
> > > Mark Brown <broonie at kernel.org> wrote:
> > 
> > > > The arch_timer case bisects to this patch in -next, regular nVHE mode
> > > > runs this test happily.
> > 
> > > My hunch is that we're missing something like the hack below, but I
> > > haven't tried it yet.
> > 
> > > I'll probably get to it tomorrow.
> > 
> > That still fails FWIW.
> 
> Yup, this has uncovered yet another pKVM bug, which doesn't preserve
> the vgic_model in its private kvm structure. I'm able to make it work
> with this:

Thanks for debugging this Marc. I've added a patch on top of kvmarm/next
with this. I don't have any A53 machines around but I was able to repro
using kvm-arm.vgic_v3_common_trap=1 on QEMU.

Thanks,
Oliver