KVM: x86: fix emulation of RSM and IRET instructions

On AMD, the effect of set_nmi_mask called by emulate_iret_real and em_rsm
on hflags is reverted later on in x86_emulate_instruction where hflags are
overwritten with ctxt->emul_flags (the kvm_set_hflags call). This manifests
as a hang when rebooting Windows VMs with QEMU, OVMF, and >1 vcpu.

Instead of trying to merge ctxt->emul_flags into vcpu->arch.hflags after
an instruction is emulated, this commit deletes emul_flags altogether and
makes the emulator access vcpu->arch.hflags using two new accessors. This
way all changes, on the emulator side as well as in functions called from
the emulator and accessing vcpu state with emul_to_vcpu, are preserved.

More details on the bug and its manifestation with Windows and OVMF:

  It's a KVM bug in the interaction between SMI/SMM and NMI, specific to AMD.
  I believe that the SMM part explains why we started seeing this only with
  OVMF.

  KVM masks and unmasks NMI when entering and leaving SMM. When KVM emulates
  the RSM instruction in em_rsm, the set_nmi_mask call doesn't stick because
  later on in x86_emulate_instruction we overwrite arch.hflags with
  ctxt->emul_flags, effectively reverting the effect of the set_nmi_mask call.
  The AMD-specific hflag of interest here is HF_NMI_MASK.

  When rebooting the system, Windows sends an NMI IPI to all but the current
  cpu to shut them down. Only after all of them are parked in HLT will the
  initiating cpu finish the restart. If NMI is masked, other cpus never get
  the memo and the initiating cpu spins forever, waiting for
  hal!HalpInterruptProcessorsStarted to drop. That's the symptom we observe.

Fixes: a584539b24b8 ("KVM: x86: pass the whole hflags field to emulator and back")
Signed-off-by: Ladi Prosek <lprosek@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

arch/x86/include/asm/kvm_emulate.h

@@ -221,6 +221,9 @@ struct x86_emulate_ops {
 	void (*get_cpuid)(struct x86_emulate_ctxt *ctxt,
 			  u32 *eax, u32 *ebx, u32 *ecx, u32 *edx);
 	void (*set_nmi_mask)(struct x86_emulate_ctxt *ctxt, bool masked);
+
+	unsigned (*get_hflags)(struct x86_emulate_ctxt *ctxt);
+	void (*set_hflags)(struct x86_emulate_ctxt *ctxt, unsigned hflags);
 };
 
 typedef u32 __attribute__((vector_size(16))) sse128_t;
@@ -290,7 +293,6 @@ struct x86_emulate_ctxt {
 
 	/* interruptibility state, as a result of execution of STI or MOV SS */
 	int interruptibility;
-	int emul_flags;
 
 	bool perm_ok; /* do not check permissions if true */
 	bool ud;	/* inject an #UD if host doesn't support insn */

arch/x86/kvm/emulate.c

@@ -2547,7 +2547,7 @@ static int em_rsm(struct x86_emulate_ctxt *ctxt)
 	u64 smbase;
 	int ret;
 
-	if ((ctxt->emul_flags & X86EMUL_SMM_MASK) == 0)
+	if ((ctxt->ops->get_hflags(ctxt) & X86EMUL_SMM_MASK) == 0)
 		return emulate_ud(ctxt);
 
 	/*
@@ -2596,11 +2596,11 @@ static int em_rsm(struct x86_emulate_ctxt *ctxt)
 		return X86EMUL_UNHANDLEABLE;
 	}
 
-	if ((ctxt->emul_flags & X86EMUL_SMM_INSIDE_NMI_MASK) == 0)
+	if ((ctxt->ops->get_hflags(ctxt) & X86EMUL_SMM_INSIDE_NMI_MASK) == 0)
 		ctxt->ops->set_nmi_mask(ctxt, false);
 
-	ctxt->emul_flags &= ~X86EMUL_SMM_INSIDE_NMI_MASK;
-	ctxt->emul_flags &= ~X86EMUL_SMM_MASK;
+	ctxt->ops->set_hflags(ctxt, ctxt->ops->get_hflags(ctxt) &
+		~(X86EMUL_SMM_INSIDE_NMI_MASK | X86EMUL_SMM_MASK));
 	return X86EMUL_CONTINUE;
 }
 
@@ -5323,6 +5323,7 @@ int x86_emulate_insn(struct x86_emulate_ctxt *ctxt)
 	const struct x86_emulate_ops *ops = ctxt->ops;
 	int rc = X86EMUL_CONTINUE;
 	int saved_dst_type = ctxt->dst.type;
+	unsigned emul_flags;
 
 	ctxt->mem_read.pos = 0;
 
@@ -5337,6 +5338,7 @@ int x86_emulate_insn(struct x86_emulate_ctxt *ctxt)
 		goto done;
 	}
 
+	emul_flags = ctxt->ops->get_hflags(ctxt);
 	if (unlikely(ctxt->d &
 		     (No64|Undefined|Sse|Mmx|Intercept|CheckPerm|Priv|Prot|String))) {
 		if ((ctxt->mode == X86EMUL_MODE_PROT64 && (ctxt->d & No64)) ||
@@ -5370,7 +5372,7 @@ int x86_emulate_insn(struct x86_emulate_ctxt *ctxt)
 				fetch_possible_mmx_operand(ctxt, &ctxt->dst);
 		}
 
-		if (unlikely(ctxt->emul_flags & X86EMUL_GUEST_MASK) && ctxt->intercept) {
+		if (unlikely(emul_flags & X86EMUL_GUEST_MASK) && ctxt->intercept) {
 			rc = emulator_check_intercept(ctxt, ctxt->intercept,
 						      X86_ICPT_PRE_EXCEPT);
 			if (rc != X86EMUL_CONTINUE)
@@ -5399,7 +5401,7 @@ int x86_emulate_insn(struct x86_emulate_ctxt *ctxt)
 				goto done;
 		}
 
-		if (unlikely(ctxt->emul_flags & X86EMUL_GUEST_MASK) && (ctxt->d & Intercept)) {
+		if (unlikely(emul_flags & X86EMUL_GUEST_MASK) && (ctxt->d & Intercept)) {
 			rc = emulator_check_intercept(ctxt, ctxt->intercept,
 						      X86_ICPT_POST_EXCEPT);
 			if (rc != X86EMUL_CONTINUE)
@@ -5453,7 +5455,7 @@ int x86_emulate_insn(struct x86_emulate_ctxt *ctxt)
 
 special_insn:
 
-	if (unlikely(ctxt->emul_flags & X86EMUL_GUEST_MASK) && (ctxt->d & Intercept)) {
+	if (unlikely(emul_flags & X86EMUL_GUEST_MASK) && (ctxt->d & Intercept)) {
 		rc = emulator_check_intercept(ctxt, ctxt->intercept,
 					      X86_ICPT_POST_MEMACCESS);
 		if (rc != X86EMUL_CONTINUE)

arch/x86/kvm/x86.c

@@ -5207,6 +5207,16 @@ static void emulator_set_nmi_mask(struct x86_emulate_ctxt *ctxt, bool masked)
 	kvm_x86_ops->set_nmi_mask(emul_to_vcpu(ctxt), masked);
 }
 
+static unsigned emulator_get_hflags(struct x86_emulate_ctxt *ctxt)
+{
+	return emul_to_vcpu(ctxt)->arch.hflags;
+}
+
+static void emulator_set_hflags(struct x86_emulate_ctxt *ctxt, unsigned emul_flags)
+{
+	kvm_set_hflags(emul_to_vcpu(ctxt), emul_flags);
+}
+
 static const struct x86_emulate_ops emulate_ops = {
 	.read_gpr            = emulator_read_gpr,
 	.write_gpr           = emulator_write_gpr,
@@ -5246,6 +5256,8 @@ static const struct x86_emulate_ops emulate_ops = {
 	.intercept           = emulator_intercept,
 	.get_cpuid           = emulator_get_cpuid,
 	.set_nmi_mask        = emulator_set_nmi_mask,
+	.get_hflags          = emulator_get_hflags,
+	.set_hflags          = emulator_set_hflags,
 };
 
 static void toggle_interruptibility(struct kvm_vcpu *vcpu, u32 mask)
@@ -5298,7 +5310,6 @@ static void init_emulate_ctxt(struct kvm_vcpu *vcpu)
 	BUILD_BUG_ON(HF_GUEST_MASK != X86EMUL_GUEST_MASK);
 	BUILD_BUG_ON(HF_SMM_MASK != X86EMUL_SMM_MASK);
 	BUILD_BUG_ON(HF_SMM_INSIDE_NMI_MASK != X86EMUL_SMM_INSIDE_NMI_MASK);
-	ctxt->emul_flags = vcpu->arch.hflags;
 
 	init_decode_cache(ctxt);
 	vcpu->arch.emulate_regs_need_sync_from_vcpu = false;
@@ -5702,8 +5713,6 @@ restart:
 		unsigned long rflags = kvm_x86_ops->get_rflags(vcpu);
 		toggle_interruptibility(vcpu, ctxt->interruptibility);
 		vcpu->arch.emulate_regs_need_sync_to_vcpu = false;
-		if (vcpu->arch.hflags != ctxt->emul_flags)
-			kvm_set_hflags(vcpu, ctxt->emul_flags);
 		kvm_rip_write(vcpu, ctxt->eip);
 		if (r == EMULATE_DONE)
 			kvm_vcpu_check_singlestep(vcpu, rflags, &r);