[PATCH 0/2] KVM: arm64: Per-vCPU TLB and I-cache flush in pKVM

[Fuad Tabba]

Hi folks,

Since 6.14, protected and non-protected pKVM guests run on the
hypervisor's VMID. The host no longer flushes a CPU's TLBs and I-cache
when a different vCPU of the same VM is loaded there, so vCPUs of one
guest can hit each other's entries. This series adds the flush at EL2.

An earlier version of these patches was in Will's pKVM base series in
2022 [1], but didn't make it into its later versions.

This series is based on kvmarm/next because the hyp VM is allocated with
hyp_alloc() there, and the per-CPU array it adds is allocated the same
way.

Based on kvmarm/next (dd0bc00e3e920).

Cheers,
/fuad

[1] https://lore.kernel.org/all/20220519134204.5379-1-will@kernel.org/

Fuad Tabba (2):
  KVM: arm64: Track physical CPU number at EL2
  KVM: arm64: Ensure TLBs and I-cache are private to each vCPU in pKVM

 arch/arm64/include/asm/kvm_hyp.h       |  3 +++
 arch/arm64/kvm/arm.c                   |  3 +++
 arch/arm64/kvm/hyp/include/nvhe/pkvm.h |  3 +++
 arch/arm64/kvm/hyp/nvhe/hyp-main.c     |  8 ++++++++
 arch/arm64/kvm/hyp/nvhe/hyp-smp.c      |  2 ++
 arch/arm64/kvm/hyp/nvhe/pkvm.c         | 21 ++++++++++++++++++---
 6 files changed, 37 insertions(+), 3 deletions(-)


base-commit: dd0bc00e3e9201fd49b62ff9cc07890c89003afe
-- 
2.39.5