Shadow paging is fundamentally incompatible with the page-modification log, because the GPAs in the log come from the wrong memory map. In particular, for the EPT page-modification log, the GPAs in the log come from L2 rather than L1. (If there was a non-EPT page-modification log, we couldn't use it for shadow paging because it would log GVAs rather than GPAs). Therefore, we need to rely on write protection to record dirty pages. This has the side effect of bypassing PML, since writes now result in an EPT violation vmexit. This turns out to be a surprisingly small patch---the testcase is what's guilty of the scary diffstat. But that is because the KVM MMU code is absurdly clever, so a very close review is appreciated. Paolo v1->v2: don't place the new code in spte_clear_dirty [Junaid] Paolo Bonzini (3): KVM: x86: assign two bits to track SPTE kinds KVM: x86: fix nested guest live migration with PML selftests: kvm: add test for dirty logging inside nested guests arch/x86/include/asm/kvm_host.h | 7 - arch/x86/kvm/mmu.c | 65 +++++-- tools/testing/selftests/kvm/Makefile | 1 + .../selftests/kvm/include/x86_64/processor.h | 3 + tools/testing/selftests/kvm/include/x86_64/vmx.h | 14 ++ tools/testing/selftests/kvm/lib/kvm_util.c | 2 +- .../testing/selftests/kvm/lib/kvm_util_internal.h | 3 + tools/testing/selftests/kvm/lib/x86_64/vmx.c | 201 ++++++++++++++++++++- .../selftests/kvm/x86_64/vmx_dirty_log_test.c | 156 ++++++++++++++++ 9 files changed, 426 insertions(+), 26 deletions(-) create mode 100644 tools/testing/selftests/kvm/x86_64/vmx_dirty_log_test.c -- 1.8.3.1
