KVM x86/mmu changes for 6.5. The new "never" option for nx_huge_pages is far and away the most interesting change. If you have an opinion on whether or not KVM should default to "never" for CPUs that aren't affected by iTLB multi-hit, now would be a wonderful time to speak up :-) https://lore.kernel.org/all/168667299355.1927151.1998349801097712999.b4-ty@xxxxxxxxxx The following changes since commit b9846a698c9aff4eb2214a06ac83638ad098f33f: KVM: VMX: add MSR_IA32_TSX_CTRL into msrs_to_save (2023-05-21 04:05:51 -0400) are available in the Git repository at: https://github.com/kvm-x86/linux.git tags/kvm-x86-mmu-6.5 for you to fetch changes up to 0b210faf337314e4bc88e796218bc70c72a51209: KVM: x86/mmu: Add "never" option to allow sticky disabling of nx_huge_pages (2023-06-13 09:16:03 -0700) ---------------------------------------------------------------- KVM x86/mmu changes for 6.5: - Add back a comment about the subtle side effect of try_cmpxchg64() in tdp_mmu_set_spte_atomic() - Add an assertion in __kvm_mmu_invalidate_addr() to verify that the target KVM MMU is the current MMU - Add a "never" option to effectively avoid creating NX hugepage recovery threads ---------------------------------------------------------------- Like Xu (1): KVM: x86/mmu: Assert on @mmu in the __kvm_mmu_invalidate_addr() Sean Christopherson (1): KVM: x86/mmu: Add "never" option to allow sticky disabling of nx_huge_pages Uros Bizjak (1): KVM: x86/mmu: Add comment on try_cmpxchg64 usage in tdp_mmu_set_spte_atomic arch/x86/kvm/mmu/mmu.c | 49 +++++++++++++++++++++++++++++++++++++++++----- arch/x86/kvm/mmu/tdp_mmu.c | 5 ++++- 2 files changed, 48 insertions(+), 6 deletions(-)
