Currently, the situation when guest accesses MMIO during vectoring is
handled differently on VMX and SVM: on VMX KVM returns internal error,
when SVM goes into infinite loop trying to deliver an event again and
again.

This patch series eliminates this difference by returning a KVM internal
error when guest performs MMIO during vectoring for both VMX and SVM.

Also, introduce a selftest test case which covers the error handling
mentioned above.

V1 -> V2:
- Make commit messages more brief, avoid using pronouns
- Extract SVM error handling into a separate commit
- Introduce a new X86EMUL_ return type and detect the unhandleable
vectoring error in vendor-specific check_emulate_instruction instead of
handling it in the common MMU code (which is specific for cached MMIO)

Ivan Orlov (6):
  KVM: x86: Add function for vectoring error generation
  KVM: x86: Add emulation status for vectoring during MMIO
  KVM: VMX: Handle vectoring error in check_emulate_instruction
  KVM: SVM: Handle MMIO during vectroing error
  selftests: KVM: extract lidt into helper function
  selftests: KVM: Add test case for MMIO during vectoring

 arch/x86/include/asm/kvm_host.h               | 12 ++++-
 arch/x86/kvm/kvm_emulate.h                    |  2 +
 arch/x86/kvm/svm/svm.c                        |  9 +++-
 arch/x86/kvm/vmx/vmx.c                        | 33 +++++-------
 arch/x86/kvm/x86.c                            | 27 ++++++++++
 .../selftests/kvm/include/x86_64/processor.h  |  7 +++
 .../selftests/kvm/set_memory_region_test.c    | 53 ++++++++++++++++++-
 .../selftests/kvm/x86_64/sev_smoke_test.c     |  2 +-
 8 files changed, 119 insertions(+), 26 deletions(-)

-- 
2.43.0