Merge tag 'arm64-fixes' of git://git.kernel.org/pub/scm/linux/kernel/…

…git/arm64/linux

Pull amr64 kvm fix from Will Deacon:
 "We've uncovered a nasty bug in the arm64 KVM code which allows a badly
  behaved 32-bit guest to bring down the host.  The fix is simple (it's
  what I believe we call a "brown paper bag" bug) and I don't think it
  makes sense to sit on this, particularly as Russell ended up
  triggering this rather than just somebody noticing a potential problem
  by inspection.

  Usually arm64 KVM changes would go via Paolo's tree, but he's on
  holiday at the moment and the deal is that anything urgent gets
  shuffled via the arch trees, so here it is.

  Summary:

  Fix arm64 KVM issue when injecting an abort into a 32-bit guest, which
  would lead to an illegal exception return at EL2 and a subsequent host
  crash"

* tag 'arm64-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux:
  arm64: KVM: Fix host crash when injecting a fault into a 32bit guest

arch/arm64/kvm/inject_fault.c

@@ -168,8 +168,8 @@ void kvm_inject_dabt(struct kvm_vcpu *vcpu, unsigned long addr)
 {
 	if (!(vcpu->arch.hcr_el2 & HCR_RW))
 		inject_abt32(vcpu, false, addr);
-
-	inject_abt64(vcpu, false, addr);
+	else
+		inject_abt64(vcpu, false, addr);
 }
 
 /**
@@ -184,8 +184,8 @@ void kvm_inject_pabt(struct kvm_vcpu *vcpu, unsigned long addr)
 {
 	if (!(vcpu->arch.hcr_el2 & HCR_RW))
 		inject_abt32(vcpu, true, addr);
-
-	inject_abt64(vcpu, true, addr);
+	else
+		inject_abt64(vcpu, true, addr);
 }
 
 /**
@@ -198,6 +198,6 @@ void kvm_inject_undefined(struct kvm_vcpu *vcpu)
 {
 	if (!(vcpu->arch.hcr_el2 & HCR_RW))
 		inject_undef32(vcpu);
-
-	inject_undef64(vcpu);
+	else
+		inject_undef64(vcpu);
 }