Merge tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm

Pull kvm fixes from Paolo Bonzini:
 "Even more x86 fixes:

   - Avoid a false positive panic in the #PF handler, if the kernel
     triggers a page fault within the asynchronous page fault handler
     before the flags are reset.

     This can happen for example because of a perf NMI.

   - Always reset a vCPU's MMU pointer back to the root MMU when loading
     nested guest state fails during RSM or KVM_SET_NESTED_STATE.

     Otherwise KVM tries to run a non-nested guest with a nested MMU,
     which can result in a NULL pointer dereference attempting to get
     the nested MMU's pgd (EPTP or nested CR3)"

* tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm:
  KVM: nVMX: Reset MMU back to the root MMU if loading L2 state fails for !VM-Entry
  x86/kvm: Don't consume async #PF flags on a kernel #PF in NMI context

2 files changed, 32 insertions, 4 deletions

diff --git a/arch/x86/kernel/kvm.c b/arch/x86/kernel/kvm.c
index 6b0a5861ccb8..d8dd2cadcbf7 100644
--- a/arch/x86/kernel/kvm.c
+++ b/arch/x86/kernel/kvm.c
@@ -265,11 +265,39 @@ noinstr u32 kvm_read_and_reset_apf_flags(void)
 }
 EXPORT_SYMBOL_FOR_KVM(kvm_read_and_reset_apf_flags);
 
+static __always_inline bool kvm_apf_reason_pending(void)
+{
+	return __this_cpu_read(async_pf_enabled) &&
+	       __this_cpu_read(apf_reason.flags);
+}
+
 noinstr bool __kvm_handle_async_pf(struct pt_regs *regs, u32 token)
 {
-	u32 flags = kvm_read_and_reset_apf_flags();
 	irqentry_state_t state;
+	u32 flags;
+
+	if (!kvm_apf_reason_pending())
+		return false;
+
+	/*
+	 * The guest does not enable KVM_ASYNC_PF_SEND_ALWAYS, so the host
+	 * delivers async #PF only to user mode (or as a #PF VM-exit, which KVM
+	 * reads itself). A #PF from kernel mode can still arrive between that
+	 * delivery and the read of the reason flags, inside an NMI that
+	 * interrupted it, for instance perf reading user memory for a call
+	 * chain. Such a fault is not async, so leave the async PF flags
+	 * untouched for the outer handler.  Outside NMI context, on the other
+	 * hand, async PF flags on a kernel-mode fault mean the host is broken.
+	 */
+	if (!user_mode(regs)) {
+		if (likely(in_nmi()))
+			return false;
+		state = irqentry_enter(regs);
+		instrumentation_begin();
+		panic("Host injected async #PF in kernel mode\n");
+	}
 
+	flags = kvm_read_and_reset_apf_flags();
 	if (!flags)
 		return false;
 
@@ -285,8 +313,6 @@ noinstr bool __kvm_handle_async_pf(struct pt_regs *regs, u32 token)
 		panic("Host injected async #PF in interrupt disabled region\n");
 
 	if (flags & KVM_PV_REASON_PAGE_NOT_PRESENT) {
-		if (unlikely(!(user_mode(regs))))
-			panic("Host injected async #PF in kernel mode\n");
 		/* Page is swapped out by the host. */
 		kvm_async_pf_task_wait_schedule(token);
 	} else {
diff --git a/arch/x86/kvm/vmx/nested.c b/arch/x86/kvm/vmx/nested.c
index b25216862740..5165275128fc 100644
--- a/arch/x86/kvm/vmx/nested.c
+++ b/arch/x86/kvm/vmx/nested.c
@@ -3770,8 +3770,10 @@ vmentry_fail_vmexit_guest_mode:
 vmentry_fail_vmexit:
 	vmx_switch_vmcs(vcpu, &vmx->vmcs01);
 
-	if (!from_vmentry)
+	if (!from_vmentry) {
+		nested_ept_uninit_mmu_context(vcpu);
 		return NVMX_VMENTRY_VMEXIT;
+	}
 
 	nested_put_vmcs12_pages(vcpu);