[v2,00/10] KVM: SVM: Attempt to cleanup SEV_FEATURES

Message

Sean Christopherson Feb. 27, 2025, 1:25 a.m. UTC

Try to address the worst of the issues that arise with guest controlled SEV
features (thanks AP creation)[1].  The most pressing issue is with DebugSwap,
as a misbehaving guest could clobber host DR masks (which should be relatively
benign?).

The other notable issue is that KVM doesn't guard against userspace manually
making a vCPU RUNNABLE after it has been DESTROYED (or after a failed CREATE).
This shouldn't be super problematic, as VMRUN is supposed to "only" fail if
the VMSA page is invalid, but passing a known bad PA to hardware isn't exactly
desirable.

[1] https://lore.kernel.org/all/[e-mail]

v2:
 - Collect reviews. [Tom, Pankaj]
 - Fix a changelog typo. [Tom]
 - Reject KVM_RUN, but don't terminate the guest if KVM attempts VMRUN
   with a bad VMSA. [Tom]
 - Fix a commment where DRs were incorreclty listed as Type-A when DebugSwap
   is disabled/unsupported. [Tom]

v1: https://lore.kernel.org/all/[e-mail]

Sean Christopherson (10):
  KVM: SVM: Save host DR masks on CPUs with DebugSwap
  KVM: SVM: Don't rely on DebugSwap to restore host DR0..DR3
  KVM: SVM: Refuse to attempt VRMUN if an SEV-ES+ guest has an invalid
    VMSA
  KVM: SVM: Don't change target vCPU state on AP Creation VMGEXIT error
  KVM: SVM: Require AP's "requested" SEV_FEATURES to match KVM's view
  KVM: SVM: Simplify request+kick logic in SNP AP Creation handling
  KVM: SVM: Use guard(mutex) to simplify SNP AP Creation error handling
  KVM: SVM: Mark VMCB dirty before processing incoming snp_vmsa_gpa
  KVM: SVM: Use guard(mutex) to simplify SNP vCPU state updates
  KVM: SVM: Invalidate "next" SNP VMSA GPA even on failure

 arch/x86/kvm/svm/sev.c | 218 +++++++++++++++++++----------------------
 arch/x86/kvm/svm/svm.c |  11 ++-
 arch/x86/kvm/svm/svm.h |   2 +-
 3 files changed, 109 insertions(+), 122 deletions(-)


base-commit: fed48e2967f402f561d80075a20c5c9e16866e53