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

Pull kvm fixes from Paolo Bonzini:
 "x86:

   - fix lock initialization race in gfn-to-pfn cache (+selftests)

   - fix two refcounting errors

   - emulator fixes

   - mask off reserved bits in CPUID

   - fix bug with disabling SGX

  RISC-V:

   - update MAINTAINERS"

* tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm:
  KVM: x86/xen: Fix eventfd error handling in kvm_xen_eventfd_assign()
  KVM: x86: smm: number of GPRs in the SMRAM image depends on the image format
  KVM: x86: emulator: update the emulation mode after CR0 write
  KVM: x86: emulator: update the emulation mode after rsm
  KVM: x86: emulator: introduce emulator_recalc_and_set_mode
  KVM: x86: emulator: em_sysexit should update ctxt->mode
  KVM: selftests: Mark "guest_saw_irq" as volatile in xen_shinfo_test
  KVM: selftests: Add tests in xen_shinfo_test to detect lock races
  KVM: Reject attempts to consume or refresh inactive gfn_to_pfn_cache
  KVM: Initialize gfn_to_pfn_cache locks in dedicated helper
  KVM: VMX: fully disable SGX if SECONDARY_EXEC_ENCLS_EXITING unavailable
  KVM: x86: Exempt pending triple fault from event injection sanity check
  MAINTAINERS: git://github -> https://github.com for kvm-riscv
  KVM: debugfs: Return retval of simple_attr_open() if it fails
  KVM: x86: Reduce refcount if single_open() fails in kvm_mmu_rmaps_stat_open()
  KVM: x86: Mask off reserved bits in CPUID.8000001FH
  KVM: x86: Mask off reserved bits in CPUID.8000001AH
  KVM: x86: Mask off reserved bits in CPUID.80000008H
  KVM: x86: Mask off reserved bits in CPUID.80000006H
  KVM: x86: Mask off reserved bits in CPUID.80000001H

MAINTAINERS

@@ -11248,7 +11248,7 @@ L:	kvm@vger.kernel.org
 L:	kvm-riscv@lists.infradead.org
 L:	linux-riscv@lists.infradead.org
 S:	Maintained
-T:	git git://github.com/kvm-riscv/linux.git
+T:	git https://github.com/kvm-riscv/linux.git
 F:	arch/riscv/include/asm/kvm*
 F:	arch/riscv/include/uapi/asm/kvm*
 F:	arch/riscv/kvm/

arch/x86/kvm/cpuid.c

@@ -1133,11 +1133,13 @@ static inline int __do_cpuid_func(struct kvm_cpuid_array *array, u32 function)
 			entry->eax = max(entry->eax, 0x80000021);
 		break;
 	case 0x80000001:
+		entry->ebx &= ~GENMASK(27, 16);
 		cpuid_entry_override(entry, CPUID_8000_0001_EDX);
 		cpuid_entry_override(entry, CPUID_8000_0001_ECX);
 		break;
 	case 0x80000006:
-		/* L2 cache and TLB: pass through host info. */
+		/* Drop reserved bits, pass host L2 cache and TLB info. */
+		entry->edx &= ~GENMASK(17, 16);
 		break;
 	case 0x80000007: /* Advanced power management */
 		/* invariant TSC is CPUID.80000007H:EDX[8] */
@@ -1167,6 +1169,7 @@ static inline int __do_cpuid_func(struct kvm_cpuid_array *array, u32 function)
 			g_phys_as = phys_as;
 
 		entry->eax = g_phys_as | (virt_as << 8);
+		entry->ecx &= ~(GENMASK(31, 16) | GENMASK(11, 8));
 		entry->edx = 0;
 		cpuid_entry_override(entry, CPUID_8000_0008_EBX);
 		break;
@@ -1186,6 +1189,9 @@ static inline int __do_cpuid_func(struct kvm_cpuid_array *array, u32 function)
 		entry->ecx = entry->edx = 0;
 		break;
 	case 0x8000001a:
+		entry->eax &= GENMASK(2, 0);
+		entry->ebx = entry->ecx = entry->edx = 0;
+		break;
 	case 0x8000001e:
 		break;
 	case 0x8000001F:
@@ -1193,7 +1199,8 @@ static inline int __do_cpuid_func(struct kvm_cpuid_array *array, u32 function)
 			entry->eax = entry->ebx = entry->ecx = entry->edx = 0;
 		} else {
 			cpuid_entry_override(entry, CPUID_8000_001F_EAX);
-
+			/* Clear NumVMPL since KVM does not support VMPL.  */
+			entry->ebx &= ~GENMASK(31, 12);
 			/*
 			 * Enumerate '0' for "PA bits reduction", the adjusted
 			 * MAXPHYADDR is enumerated directly (see 0x80000008).

arch/x86/kvm/debugfs.c

@@ -158,11 +158,16 @@ static int kvm_mmu_rmaps_stat_show(struct seq_file *m, void *v)
 static int kvm_mmu_rmaps_stat_open(struct inode *inode, struct file *file)
 {
 	struct kvm *kvm = inode->i_private;
+	int r;
 
 	if (!kvm_get_kvm_safe(kvm))
 		return -ENOENT;
 
-	return single_open(file, kvm_mmu_rmaps_stat_show, kvm);
+	r = single_open(file, kvm_mmu_rmaps_stat_show, kvm);
+	if (r < 0)
+		kvm_put_kvm(kvm);
+
+	return r;
 }
 
 static int kvm_mmu_rmaps_stat_release(struct inode *inode, struct file *file)

arch/x86/kvm/emulate.c

@@ -791,8 +791,7 @@ static int linearize(struct x86_emulate_ctxt *ctxt,
 			   ctxt->mode, linear);
 }
 
-static inline int assign_eip(struct x86_emulate_ctxt *ctxt, ulong dst,
-			     enum x86emul_mode mode)
+static inline int assign_eip(struct x86_emulate_ctxt *ctxt, ulong dst)
 {
 	ulong linear;
 	int rc;
@@ -802,41 +801,71 @@ static inline int assign_eip(struct x86_emulate_ctxt *ctxt, ulong dst,
 
 	if (ctxt->op_bytes != sizeof(unsigned long))
 		addr.ea = dst & ((1UL << (ctxt->op_bytes << 3)) - 1);
-	rc = __linearize(ctxt, addr, &max_size, 1, false, true, mode, &linear);
+	rc = __linearize(ctxt, addr, &max_size, 1, false, true, ctxt->mode, &linear);
 	if (rc == X86EMUL_CONTINUE)
 		ctxt->_eip = addr.ea;
 	return rc;
 }
 
-static inline int assign_eip_near(struct x86_emulate_ctxt *ctxt, ulong dst)
+static inline int emulator_recalc_and_set_mode(struct x86_emulate_ctxt *ctxt)
 {
-	return assign_eip(ctxt, dst, ctxt->mode);
-}
+	u64 efer;
+	struct desc_struct cs;
+	u16 selector;
+	u32 base3;
 
-static int assign_eip_far(struct x86_emulate_ctxt *ctxt, ulong dst,
-			  const struct desc_struct *cs_desc)
-{
-	enum x86emul_mode mode = ctxt->mode;
-	int rc;
+	ctxt->ops->get_msr(ctxt, MSR_EFER, &efer);
 
-#ifdef CONFIG_X86_64
-	if (ctxt->mode >= X86EMUL_MODE_PROT16) {
-		if (cs_desc->l) {
-			u64 efer = 0;
+	if (!(ctxt->ops->get_cr(ctxt, 0) & X86_CR0_PE)) {
+		/* Real mode. cpu must not have long mode active */
+		if (efer & EFER_LMA)
+			return X86EMUL_UNHANDLEABLE;
+		ctxt->mode = X86EMUL_MODE_REAL;
+		return X86EMUL_CONTINUE;
+	}
 
-			ctxt->ops->get_msr(ctxt, MSR_EFER, &efer);
+	if (ctxt->eflags & X86_EFLAGS_VM) {
+		/* Protected/VM86 mode. cpu must not have long mode active */
 		if (efer & EFER_LMA)
-				mode = X86EMUL_MODE_PROT64;
-		} else
-			mode = X86EMUL_MODE_PROT32; /* temporary value */
+			return X86EMUL_UNHANDLEABLE;
+		ctxt->mode = X86EMUL_MODE_VM86;
+		return X86EMUL_CONTINUE;
 	}
-#endif
-	if (mode == X86EMUL_MODE_PROT16 || mode == X86EMUL_MODE_PROT32)
-		mode = cs_desc->d ? X86EMUL_MODE_PROT32 : X86EMUL_MODE_PROT16;
-	rc = assign_eip(ctxt, dst, mode);
-	if (rc == X86EMUL_CONTINUE)
-		ctxt->mode = mode;
+
+	if (!ctxt->ops->get_segment(ctxt, &selector, &cs, &base3, VCPU_SREG_CS))
+		return X86EMUL_UNHANDLEABLE;
+
+	if (efer & EFER_LMA) {
+		if (cs.l) {
+			/* Proper long mode */
+			ctxt->mode = X86EMUL_MODE_PROT64;
+		} else if (cs.d) {
+			/* 32 bit compatibility mode*/
+			ctxt->mode = X86EMUL_MODE_PROT32;
+		} else {
+			ctxt->mode = X86EMUL_MODE_PROT16;
+		}
+	} else {
+		/* Legacy 32 bit / 16 bit mode */
+		ctxt->mode = cs.d ? X86EMUL_MODE_PROT32 : X86EMUL_MODE_PROT16;
+	}
+
+	return X86EMUL_CONTINUE;
+}
+
+static inline int assign_eip_near(struct x86_emulate_ctxt *ctxt, ulong dst)
+{
+	return assign_eip(ctxt, dst);
+}
+
+static int assign_eip_far(struct x86_emulate_ctxt *ctxt, ulong dst)
+{
+	int rc = emulator_recalc_and_set_mode(ctxt);
+
+	if (rc != X86EMUL_CONTINUE)
 		return rc;
+
+	return assign_eip(ctxt, dst);
 }
 
 static inline int jmp_rel(struct x86_emulate_ctxt *ctxt, int rel)
@@ -2172,7 +2201,7 @@ static int em_jmp_far(struct x86_emulate_ctxt *ctxt)
 	if (rc != X86EMUL_CONTINUE)
 		return rc;
 
-	rc = assign_eip_far(ctxt, ctxt->src.val, &new_desc);
+	rc = assign_eip_far(ctxt, ctxt->src.val);
 	/* Error handling is not implemented. */
 	if (rc != X86EMUL_CONTINUE)
 		return X86EMUL_UNHANDLEABLE;
@@ -2250,7 +2279,7 @@ static int em_ret_far(struct x86_emulate_ctxt *ctxt)
 				       &new_desc);
 	if (rc != X86EMUL_CONTINUE)
 		return rc;
-	rc = assign_eip_far(ctxt, eip, &new_desc);
+	rc = assign_eip_far(ctxt, eip);
 	/* Error handling is not implemented. */
 	if (rc != X86EMUL_CONTINUE)
 		return X86EMUL_UNHANDLEABLE;
@@ -2432,7 +2461,7 @@ static int rsm_load_state_32(struct x86_emulate_ctxt *ctxt,
 	ctxt->eflags =             GET_SMSTATE(u32, smstate, 0x7ff4) | X86_EFLAGS_FIXED;
 	ctxt->_eip =               GET_SMSTATE(u32, smstate, 0x7ff0);
 
-	for (i = 0; i < NR_EMULATOR_GPRS; i++)
+	for (i = 0; i < 8; i++)
 		*reg_write(ctxt, i) = GET_SMSTATE(u32, smstate, 0x7fd0 + i * 4);
 
 	val = GET_SMSTATE(u32, smstate, 0x7fcc);
@@ -2489,7 +2518,7 @@ static int rsm_load_state_64(struct x86_emulate_ctxt *ctxt,
 	u16 selector;
 	int i, r;
 
-	for (i = 0; i < NR_EMULATOR_GPRS; i++)
+	for (i = 0; i < 16; i++)
 		*reg_write(ctxt, i) = GET_SMSTATE(u64, smstate, 0x7ff8 - i * 8);
 
 	ctxt->_eip   = GET_SMSTATE(u64, smstate, 0x7f78);
@@ -2633,7 +2662,7 @@ static int em_rsm(struct x86_emulate_ctxt *ctxt)
 	 * those side effects need to be explicitly handled for both success
 	 * and shutdown.
 	 */
-	return X86EMUL_CONTINUE;
+	return emulator_recalc_and_set_mode(ctxt);
 
 emulate_shutdown:
 	ctxt->ops->triple_fault(ctxt);
@@ -2876,6 +2905,7 @@ static int em_sysexit(struct x86_emulate_ctxt *ctxt)
 	ops->set_segment(ctxt, ss_sel, &ss, 0, VCPU_SREG_SS);
 
 	ctxt->_eip = rdx;
+	ctxt->mode = usermode;
 	*reg_write(ctxt, VCPU_REGS_RSP) = rcx;
 
 	return X86EMUL_CONTINUE;
@@ -3469,7 +3499,7 @@ static int em_call_far(struct x86_emulate_ctxt *ctxt)
 	if (rc != X86EMUL_CONTINUE)
 		return rc;
 
-	rc = assign_eip_far(ctxt, ctxt->src.val, &new_desc);
+	rc = assign_eip_far(ctxt, ctxt->src.val);
 	if (rc != X86EMUL_CONTINUE)
 		goto fail;
 
@@ -3611,11 +3641,25 @@ static int em_movbe(struct x86_emulate_ctxt *ctxt)
 
 static int em_cr_write(struct x86_emulate_ctxt *ctxt)
 {
-	if (ctxt->ops->set_cr(ctxt, ctxt->modrm_reg, ctxt->src.val))
+	int cr_num = ctxt->modrm_reg;
+	int r;
+
+	if (ctxt->ops->set_cr(ctxt, cr_num, ctxt->src.val))
 		return emulate_gp(ctxt, 0);
 
 	/* Disable writeback. */
 	ctxt->dst.type = OP_NONE;
+
+	if (cr_num == 0) {
+		/*
+		 * CR0 write might have updated CR0.PE and/or CR0.PG
+		 * which can affect the cpu's execution mode.
+		 */
+		r = emulator_recalc_and_set_mode(ctxt);
+		if (r != X86EMUL_CONTINUE)
+			return r;
+	}
+
 	return X86EMUL_CONTINUE;
 }
 

arch/x86/kvm/vmx/vmx.c

@@ -8263,6 +8263,11 @@ static __init int hardware_setup(void)
 	if (!cpu_has_virtual_nmis())
 		enable_vnmi = 0;
 
+#ifdef CONFIG_X86_SGX_KVM
+	if (!cpu_has_vmx_encls_vmexit())
+		enable_sgx = false;
+#endif
+
 	/*
 	 * set_apic_access_page_addr() is used to reload apic access
 	 * page upon invalidation.  No need to do anything if not

arch/x86/kvm/x86.c

@@ -2315,11 +2315,11 @@ static void kvm_write_system_time(struct kvm_vcpu *vcpu, gpa_t system_time,
 
 	/* we verify if the enable bit is set... */
 	if (system_time & 1) {
-		kvm_gfn_to_pfn_cache_init(vcpu->kvm, &vcpu->arch.pv_time, vcpu,
+		kvm_gpc_activate(vcpu->kvm, &vcpu->arch.pv_time, vcpu,
 				 KVM_HOST_USES_PFN, system_time & ~1ULL,
 				 sizeof(struct pvclock_vcpu_time_info));
 	} else {
-		kvm_gfn_to_pfn_cache_destroy(vcpu->kvm, &vcpu->arch.pv_time);
+		kvm_gpc_deactivate(vcpu->kvm, &vcpu->arch.pv_time);
 	}
 
 	return;
@@ -3388,7 +3388,7 @@ static int kvm_pv_enable_async_pf_int(struct kvm_vcpu *vcpu, u64 data)
 
 static void kvmclock_reset(struct kvm_vcpu *vcpu)
 {
-	kvm_gfn_to_pfn_cache_destroy(vcpu->kvm, &vcpu->arch.pv_time);
+	kvm_gpc_deactivate(vcpu->kvm, &vcpu->arch.pv_time);
 	vcpu->arch.time = 0;
 }
 
@@ -10044,7 +10044,20 @@ static int kvm_check_and_inject_events(struct kvm_vcpu *vcpu,
 	    kvm_x86_ops.nested_ops->has_events(vcpu))
 		*req_immediate_exit = true;
 
-	WARN_ON(kvm_is_exception_pending(vcpu));
+	/*
+	 * KVM must never queue a new exception while injecting an event; KVM
+	 * is done emulating and should only propagate the to-be-injected event
+	 * to the VMCS/VMCB.  Queueing a new exception can put the vCPU into an
+	 * infinite loop as KVM will bail from VM-Enter to inject the pending
+	 * exception and start the cycle all over.
+	 *
+	 * Exempt triple faults as they have special handling and won't put the
+	 * vCPU into an infinite loop.  Triple fault can be queued when running
+	 * VMX without unrestricted guest, as that requires KVM to emulate Real
+	 * Mode events (see kvm_inject_realmode_interrupt()).
+	 */
+	WARN_ON_ONCE(vcpu->arch.exception.pending ||
+		     vcpu->arch.exception_vmexit.pending);
 	return 0;
 
 out:
@@ -11816,6 +11829,8 @@ int kvm_arch_vcpu_create(struct kvm_vcpu *vcpu)
 	vcpu->arch.regs_avail = ~0;
 	vcpu->arch.regs_dirty = ~0;
 
+	kvm_gpc_init(&vcpu->arch.pv_time);
+
 	if (!irqchip_in_kernel(vcpu->kvm) || kvm_vcpu_is_reset_bsp(vcpu))
 		vcpu->arch.mp_state = KVM_MP_STATE_RUNNABLE;
 	else

arch/x86/kvm/xen.c

@@ -42,13 +42,13 @@ static int kvm_xen_shared_info_init(struct kvm *kvm, gfn_t gfn)
 	int idx = srcu_read_lock(&kvm->srcu);
 
 	if (gfn == GPA_INVALID) {
-		kvm_gfn_to_pfn_cache_destroy(kvm, gpc);
+		kvm_gpc_deactivate(kvm, gpc);
 		goto out;
 	}
 
 	do {
-		ret = kvm_gfn_to_pfn_cache_init(kvm, gpc, NULL, KVM_HOST_USES_PFN,
-						gpa, PAGE_SIZE);
+		ret = kvm_gpc_activate(kvm, gpc, NULL, KVM_HOST_USES_PFN, gpa,
+				       PAGE_SIZE);
 		if (ret)
 			goto out;
 
@@ -554,14 +554,14 @@ int kvm_xen_vcpu_set_attr(struct kvm_vcpu *vcpu, struct kvm_xen_vcpu_attr *data)
 			     offsetof(struct compat_vcpu_info, time));
 
 		if (data->u.gpa == GPA_INVALID) {
-			kvm_gfn_to_pfn_cache_destroy(vcpu->kvm, &vcpu->arch.xen.vcpu_info_cache);
+			kvm_gpc_deactivate(vcpu->kvm, &vcpu->arch.xen.vcpu_info_cache);
 			r = 0;
 			break;
 		}
 
-		r = kvm_gfn_to_pfn_cache_init(vcpu->kvm,
-					      &vcpu->arch.xen.vcpu_info_cache,
-					      NULL, KVM_HOST_USES_PFN, data->u.gpa,
+		r = kvm_gpc_activate(vcpu->kvm,
+				     &vcpu->arch.xen.vcpu_info_cache, NULL,
+				     KVM_HOST_USES_PFN, data->u.gpa,
 				     sizeof(struct vcpu_info));
 		if (!r)
 			kvm_make_request(KVM_REQ_CLOCK_UPDATE, vcpu);
@@ -570,13 +570,13 @@ int kvm_xen_vcpu_set_attr(struct kvm_vcpu *vcpu, struct kvm_xen_vcpu_attr *data)
 
 	case KVM_XEN_VCPU_ATTR_TYPE_VCPU_TIME_INFO:
 		if (data->u.gpa == GPA_INVALID) {
-			kvm_gfn_to_pfn_cache_destroy(vcpu->kvm,
+			kvm_gpc_deactivate(vcpu->kvm,
 					   &vcpu->arch.xen.vcpu_time_info_cache);
 			r = 0;
 			break;
 		}
 
-		r = kvm_gfn_to_pfn_cache_init(vcpu->kvm,
+		r = kvm_gpc_activate(vcpu->kvm,
 				     &vcpu->arch.xen.vcpu_time_info_cache,
 				     NULL, KVM_HOST_USES_PFN, data->u.gpa,
 				     sizeof(struct pvclock_vcpu_time_info));
@@ -590,14 +590,13 @@ int kvm_xen_vcpu_set_attr(struct kvm_vcpu *vcpu, struct kvm_xen_vcpu_attr *data)
 			break;
 		}
 		if (data->u.gpa == GPA_INVALID) {
-			kvm_gfn_to_pfn_cache_destroy(vcpu->kvm,
+			kvm_gpc_deactivate(vcpu->kvm,
 					   &vcpu->arch.xen.runstate_cache);
 			r = 0;
 			break;
 		}
 
-		r = kvm_gfn_to_pfn_cache_init(vcpu->kvm,
-					      &vcpu->arch.xen.runstate_cache,
+		r = kvm_gpc_activate(vcpu->kvm, &vcpu->arch.xen.runstate_cache,
 				     NULL, KVM_HOST_USES_PFN, data->u.gpa,
 				     sizeof(struct vcpu_runstate_info));
 		break;
@@ -1667,18 +1666,18 @@ static int kvm_xen_eventfd_assign(struct kvm *kvm,
 	case EVTCHNSTAT_ipi:
 		/* IPI  must map back to the same port# */
 		if (data->u.evtchn.deliver.port.port != data->u.evtchn.send_port)
-			goto out; /* -EINVAL */
+			goto out_noeventfd; /* -EINVAL */
 		break;
 
 	case EVTCHNSTAT_interdomain:
 		if (data->u.evtchn.deliver.port.port) {
 			if (data->u.evtchn.deliver.port.port >= max_evtchn_port(kvm))
-				goto out; /* -EINVAL */
+				goto out_noeventfd; /* -EINVAL */
 		} else {
 			eventfd = eventfd_ctx_fdget(data->u.evtchn.deliver.eventfd.fd);
 			if (IS_ERR(eventfd)) {
 				ret = PTR_ERR(eventfd);
-				goto out;
+				goto out_noeventfd;
 			}
 		}
 		break;
@@ -1718,6 +1717,7 @@ static int kvm_xen_eventfd_assign(struct kvm *kvm,
 out:
 	if (eventfd)
 		eventfd_ctx_put(eventfd);
+out_noeventfd:
 	kfree(evtchnfd);
 	return ret;
 }
@@ -1816,7 +1816,12 @@ void kvm_xen_init_vcpu(struct kvm_vcpu *vcpu)
 {
 	vcpu->arch.xen.vcpu_id = vcpu->vcpu_idx;
 	vcpu->arch.xen.poll_evtchn = 0;
+
 	timer_setup(&vcpu->arch.xen.poll_timer, cancel_evtchn_poll, 0);
+
+	kvm_gpc_init(&vcpu->arch.xen.runstate_cache);
+	kvm_gpc_init(&vcpu->arch.xen.vcpu_info_cache);
+	kvm_gpc_init(&vcpu->arch.xen.vcpu_time_info_cache);
 }
 
 void kvm_xen_destroy_vcpu(struct kvm_vcpu *vcpu)
@@ -1824,18 +1829,17 @@ void kvm_xen_destroy_vcpu(struct kvm_vcpu *vcpu)
 	if (kvm_xen_timer_enabled(vcpu))
 		kvm_xen_stop_timer(vcpu);
 
-	kvm_gfn_to_pfn_cache_destroy(vcpu->kvm,
-				     &vcpu->arch.xen.runstate_cache);
-	kvm_gfn_to_pfn_cache_destroy(vcpu->kvm,
-				     &vcpu->arch.xen.vcpu_info_cache);
-	kvm_gfn_to_pfn_cache_destroy(vcpu->kvm,
-				     &vcpu->arch.xen.vcpu_time_info_cache);
+	kvm_gpc_deactivate(vcpu->kvm, &vcpu->arch.xen.runstate_cache);
+	kvm_gpc_deactivate(vcpu->kvm, &vcpu->arch.xen.vcpu_info_cache);
+	kvm_gpc_deactivate(vcpu->kvm, &vcpu->arch.xen.vcpu_time_info_cache);
+
 	del_timer_sync(&vcpu->arch.xen.poll_timer);
 }
 
 void kvm_xen_init_vm(struct kvm *kvm)
 {
 	idr_init(&kvm->arch.xen.evtchn_ports);
+	kvm_gpc_init(&kvm->arch.xen.shinfo_cache);
 }
 
 void kvm_xen_destroy_vm(struct kvm *kvm)
@@ -1843,7 +1847,7 @@ void kvm_xen_destroy_vm(struct kvm *kvm)
 	struct evtchnfd *evtchnfd;
 	int i;
 
-	kvm_gfn_to_pfn_cache_destroy(kvm, &kvm->arch.xen.shinfo_cache);
+	kvm_gpc_deactivate(kvm, &kvm->arch.xen.shinfo_cache);
 
 	idr_for_each_entry(&kvm->arch.xen.evtchn_ports, evtchnfd, i) {
 		if (!evtchnfd->deliver.port.port)

include/linux/kvm_host.h

@@ -1240,8 +1240,18 @@ int kvm_vcpu_write_guest(struct kvm_vcpu *vcpu, gpa_t gpa, const void *data,
 void kvm_vcpu_mark_page_dirty(struct kvm_vcpu *vcpu, gfn_t gfn);
 
 /**
- * kvm_gfn_to_pfn_cache_init - prepare a cached kernel mapping and HPA for a
- *                             given guest physical address.
+ * kvm_gpc_init - initialize gfn_to_pfn_cache.
+ *
+ * @gpc:	   struct gfn_to_pfn_cache object.
+ *
+ * This sets up a gfn_to_pfn_cache by initializing locks.  Note, the cache must
+ * be zero-allocated (or zeroed by the caller before init).
+ */
+void kvm_gpc_init(struct gfn_to_pfn_cache *gpc);
+
+/**
+ * kvm_gpc_activate - prepare a cached kernel mapping and HPA for a given guest
+ *                    physical address.
  *
  * @kvm:	   pointer to kvm instance.
  * @gpc:	   struct gfn_to_pfn_cache object.
@@ -1265,7 +1275,7 @@ void kvm_vcpu_mark_page_dirty(struct kvm_vcpu *vcpu, gfn_t gfn);
  * kvm_gfn_to_pfn_cache_check() to ensure that the cache is valid before
  * accessing the target page.
  */
-int kvm_gfn_to_pfn_cache_init(struct kvm *kvm, struct gfn_to_pfn_cache *gpc,
+int kvm_gpc_activate(struct kvm *kvm, struct gfn_to_pfn_cache *gpc,
 		     struct kvm_vcpu *vcpu, enum pfn_cache_usage usage,
 		     gpa_t gpa, unsigned long len);
 
@@ -1324,7 +1334,7 @@ int kvm_gfn_to_pfn_cache_refresh(struct kvm *kvm, struct gfn_to_pfn_cache *gpc,
 void kvm_gfn_to_pfn_cache_unmap(struct kvm *kvm, struct gfn_to_pfn_cache *gpc);
 
 /**
- * kvm_gfn_to_pfn_cache_destroy - destroy and unlink a gfn_to_pfn_cache.
+ * kvm_gpc_deactivate - deactivate and unlink a gfn_to_pfn_cache.
  *
  * @kvm:	   pointer to kvm instance.
  * @gpc:	   struct gfn_to_pfn_cache object.
@@ -1332,7 +1342,7 @@ void kvm_gfn_to_pfn_cache_unmap(struct kvm *kvm, struct gfn_to_pfn_cache *gpc);
  * This removes a cache from the @kvm's list to be processed on MMU notifier
  * invocation.
  */
-void kvm_gfn_to_pfn_cache_destroy(struct kvm *kvm, struct gfn_to_pfn_cache *gpc);
+void kvm_gpc_deactivate(struct kvm *kvm, struct gfn_to_pfn_cache *gpc);
 
 void kvm_sigset_activate(struct kvm_vcpu *vcpu);
 void kvm_sigset_deactivate(struct kvm_vcpu *vcpu);

tools/testing/selftests/kvm/x86_64/xen_shinfo_test.c

@@ -15,9 +15,13 @@
 #include <time.h>
 #include <sched.h>
 #include <signal.h>
+#include <pthread.h>
 
 #include <sys/eventfd.h>
 
+/* Defined in include/linux/kvm_types.h */
+#define GPA_INVALID		(~(ulong)0)
+
 #define SHINFO_REGION_GVA	0xc0000000ULL
 #define SHINFO_REGION_GPA	0xc0000000ULL
 #define SHINFO_REGION_SLOT	10
@@ -44,6 +48,8 @@
 
 #define MIN_STEAL_TIME		50000
 
+#define SHINFO_RACE_TIMEOUT	2	/* seconds */
+
 #define __HYPERVISOR_set_timer_op	15
 #define __HYPERVISOR_sched_op		29
 #define __HYPERVISOR_event_channel_op	32
@@ -126,7 +132,7 @@ struct {
 	struct kvm_irq_routing_entry entries[2];
 } irq_routes;
 
-bool guest_saw_irq;
+static volatile bool guest_saw_irq;
 
 static void evtchn_handler(struct ex_regs *regs)
 {
@@ -148,6 +154,7 @@ static void guest_wait_for_irq(void)
 static void guest_code(void)
 {
 	struct vcpu_runstate_info *rs = (void *)RUNSTATE_VADDR;
+	int i;
 
 	__asm__ __volatile__(
 		"sti\n"
@@ -325,6 +332,49 @@ static void guest_code(void)
 	guest_wait_for_irq();
 
 	GUEST_SYNC(21);
+	/* Racing host ioctls */
+
+	guest_wait_for_irq();
+
+	GUEST_SYNC(22);
+	/* Racing vmcall against host ioctl */
+
+	ports[0] = 0;
+
+	p = (struct sched_poll) {
+		.ports = ports,
+		.nr_ports = 1,
+		.timeout = 0
+	};
+
+wait_for_timer:
+	/*
+	 * Poll for a timer wake event while the worker thread is mucking with
+	 * the shared info.  KVM XEN drops timer IRQs if the shared info is
+	 * invalid when the timer expires.  Arbitrarily poll 100 times before
+	 * giving up and asking the VMM to re-arm the timer.  100 polls should
+	 * consume enough time to beat on KVM without taking too long if the
+	 * timer IRQ is dropped due to an invalid event channel.
+	 */
+	for (i = 0; i < 100 && !guest_saw_irq; i++)
+		asm volatile("vmcall"
+			     : "=a" (rax)
+			     : "a" (__HYPERVISOR_sched_op),
+			       "D" (SCHEDOP_poll),
+			       "S" (&p)
+			     : "memory");
+
+	/*
+	 * Re-send the timer IRQ if it was (likely) dropped due to the timer
+	 * expiring while the event channel was invalid.
+	 */
+	if (!guest_saw_irq) {
+		GUEST_SYNC(23);
+		goto wait_for_timer;
+	}
+	guest_saw_irq = false;
+
+	GUEST_SYNC(24);
 }
 
 static int cmp_timespec(struct timespec *a, struct timespec *b)
@@ -352,11 +402,36 @@ static void handle_alrm(int sig)
 	TEST_FAIL("IRQ delivery timed out");
 }
 
+static void *juggle_shinfo_state(void *arg)
+{
+	struct kvm_vm *vm = (struct kvm_vm *)arg;
+
+	struct kvm_xen_hvm_attr cache_init = {
+		.type = KVM_XEN_ATTR_TYPE_SHARED_INFO,
+		.u.shared_info.gfn = SHINFO_REGION_GPA / PAGE_SIZE
+	};
+
+	struct kvm_xen_hvm_attr cache_destroy = {
+		.type = KVM_XEN_ATTR_TYPE_SHARED_INFO,
+		.u.shared_info.gfn = GPA_INVALID
+	};
+
+	for (;;) {
+		__vm_ioctl(vm, KVM_XEN_HVM_SET_ATTR, &cache_init);
+		__vm_ioctl(vm, KVM_XEN_HVM_SET_ATTR, &cache_destroy);
+		pthread_testcancel();
+	};
+
+	return NULL;
+}
+
 int main(int argc, char *argv[])
 {
 	struct timespec min_ts, max_ts, vm_ts;
 	struct kvm_vm *vm;
+	pthread_t thread;
 	bool verbose;
+	int ret;
 
 	verbose = argc > 1 && (!strncmp(argv[1], "-v", 3) ||
 			       !strncmp(argv[1], "--verbose", 10));
@@ -785,6 +860,71 @@ int main(int argc, char *argv[])
 			case 21:
 				TEST_ASSERT(!evtchn_irq_expected,
 					    "Expected event channel IRQ but it didn't happen");
+				alarm(0);
+
+				if (verbose)
+					printf("Testing shinfo lock corruption (KVM_XEN_HVM_EVTCHN_SEND)\n");
+
+				ret = pthread_create(&thread, NULL, &juggle_shinfo_state, (void *)vm);
+				TEST_ASSERT(ret == 0, "pthread_create() failed: %s", strerror(ret));
+
+				struct kvm_irq_routing_xen_evtchn uxe = {
+					.port = 1,
+					.vcpu = vcpu->id,
+					.priority = KVM_IRQ_ROUTING_XEN_EVTCHN_PRIO_2LEVEL
+				};
+
+				evtchn_irq_expected = true;
+				for (time_t t = time(NULL) + SHINFO_RACE_TIMEOUT; time(NULL) < t;)
+					__vm_ioctl(vm, KVM_XEN_HVM_EVTCHN_SEND, &uxe);
+				break;
+
+			case 22:
+				TEST_ASSERT(!evtchn_irq_expected,
+					    "Expected event channel IRQ but it didn't happen");
+
+				if (verbose)
+					printf("Testing shinfo lock corruption (SCHEDOP_poll)\n");
+
+				shinfo->evtchn_pending[0] = 1;
+
+				evtchn_irq_expected = true;
+				tmr.u.timer.expires_ns = rs->state_entry_time +
+							 SHINFO_RACE_TIMEOUT * 1000000000ULL;
+				vcpu_ioctl(vcpu, KVM_XEN_VCPU_SET_ATTR, &tmr);
+				break;
+
+			case 23:
+				/*
+				 * Optional and possibly repeated sync point.
+				 * Injecting the timer IRQ may fail if the
+				 * shinfo is invalid when the timer expires.
+				 * If the timer has expired but the IRQ hasn't
+				 * been delivered, rearm the timer and retry.
+				 */
+				vcpu_ioctl(vcpu, KVM_XEN_VCPU_GET_ATTR, &tmr);
+
+				/* Resume the guest if the timer is still pending. */
+				if (tmr.u.timer.expires_ns)
+					break;
+
+				/* All done if the IRQ was delivered. */
+				if (!evtchn_irq_expected)
+					break;
+
+				tmr.u.timer.expires_ns = rs->state_entry_time +
+							 SHINFO_RACE_TIMEOUT * 1000000000ULL;
+				vcpu_ioctl(vcpu, KVM_XEN_VCPU_SET_ATTR, &tmr);
+				break;
+			case 24:
+				TEST_ASSERT(!evtchn_irq_expected,
+					    "Expected event channel IRQ but it didn't happen");
+
+				ret = pthread_cancel(thread);
+				TEST_ASSERT(ret == 0, "pthread_cancel() failed: %s", strerror(ret));
+
+				ret = pthread_join(thread, 0);
+				TEST_ASSERT(ret == 0, "pthread_join() failed: %s", strerror(ret));
 				goto done;
 
 			case 0x20:

virt/kvm/kvm_main.c

@@ -5409,6 +5409,7 @@ static int kvm_debugfs_open(struct inode *inode, struct file *file,
 			   int (*get)(void *, u64 *), int (*set)(void *, u64),
 			   const char *fmt)
 {
+	int ret;
 	struct kvm_stat_data *stat_data = (struct kvm_stat_data *)
 					  inode->i_private;
 
@@ -5420,15 +5421,13 @@ static int kvm_debugfs_open(struct inode *inode, struct file *file,
 	if (!kvm_get_kvm_safe(stat_data->kvm))
 		return -ENOENT;
 
-	if (simple_attr_open(inode, file, get,
+	ret = simple_attr_open(inode, file, get,
 			       kvm_stats_debugfs_mode(stat_data->desc) & 0222
-		    ? set : NULL,
-		    fmt)) {
+			       ? set : NULL, fmt);
+	if (ret)
 		kvm_put_kvm(stat_data->kvm);
-		return -ENOMEM;
-	}
 
-	return 0;
+	return ret;
 }
 
 static int kvm_debugfs_release(struct inode *inode, struct file *file)

virt/kvm/pfncache.c

@@ -81,6 +81,9 @@ bool kvm_gfn_to_pfn_cache_check(struct kvm *kvm, struct gfn_to_pfn_cache *gpc,
 {
 	struct kvm_memslots *slots = kvm_memslots(kvm);
 
+	if (!gpc->active)
+		return false;
+
 	if ((gpa & ~PAGE_MASK) + len > PAGE_SIZE)
 		return false;
 
@@ -240,10 +243,11 @@ int kvm_gfn_to_pfn_cache_refresh(struct kvm *kvm, struct gfn_to_pfn_cache *gpc,
 {
 	struct kvm_memslots *slots = kvm_memslots(kvm);
 	unsigned long page_offset = gpa & ~PAGE_MASK;
-	kvm_pfn_t old_pfn, new_pfn;
+	bool unmap_old = false;
 	unsigned long old_uhva;
+	kvm_pfn_t old_pfn;
 	void *old_khva;
-	int ret = 0;
+	int ret;
 
 	/*
 	 * If must fit within a single page. The 'len' argument is
@@ -261,6 +265,11 @@ int kvm_gfn_to_pfn_cache_refresh(struct kvm *kvm, struct gfn_to_pfn_cache *gpc,
 
 	write_lock_irq(&gpc->lock);
 
+	if (!gpc->active) {
+		ret = -EINVAL;
+		goto out_unlock;
+	}
+
 	old_pfn = gpc->pfn;
 	old_khva = gpc->khva - offset_in_page(gpc->khva);
 	old_uhva = gpc->uhva;
@@ -291,6 +300,7 @@ int kvm_gfn_to_pfn_cache_refresh(struct kvm *kvm, struct gfn_to_pfn_cache *gpc,
 		/* If the HVA→PFN mapping was already valid, don't unmap it. */
 		old_pfn = KVM_PFN_ERR_FAULT;
 		old_khva = NULL;
+		ret = 0;
 	}
 
  out:
@@ -305,14 +315,15 @@ int kvm_gfn_to_pfn_cache_refresh(struct kvm *kvm, struct gfn_to_pfn_cache *gpc,
 		gpc->khva = NULL;
 	}
 
-	/* Snapshot the new pfn before dropping the lock! */
-	new_pfn = gpc->pfn;
+	/* Detect a pfn change before dropping the lock! */
+	unmap_old = (old_pfn != gpc->pfn);
 
+out_unlock:
 	write_unlock_irq(&gpc->lock);
 
 	mutex_unlock(&gpc->refresh_lock);
 
-	if (old_pfn != new_pfn)
+	if (unmap_old)
 		gpc_unmap_khva(kvm, old_pfn, old_khva);
 
 	return ret;
@@ -346,42 +357,61 @@ void kvm_gfn_to_pfn_cache_unmap(struct kvm *kvm, struct gfn_to_pfn_cache *gpc)
 }
 EXPORT_SYMBOL_GPL(kvm_gfn_to_pfn_cache_unmap);
 
+void kvm_gpc_init(struct gfn_to_pfn_cache *gpc)
+{
+	rwlock_init(&gpc->lock);
+	mutex_init(&gpc->refresh_lock);
+}
+EXPORT_SYMBOL_GPL(kvm_gpc_init);
 
-int kvm_gfn_to_pfn_cache_init(struct kvm *kvm, struct gfn_to_pfn_cache *gpc,
+int kvm_gpc_activate(struct kvm *kvm, struct gfn_to_pfn_cache *gpc,
 		     struct kvm_vcpu *vcpu, enum pfn_cache_usage usage,
 		     gpa_t gpa, unsigned long len)
 {
 	WARN_ON_ONCE(!usage || (usage & KVM_GUEST_AND_HOST_USE_PFN) != usage);
 
 	if (!gpc->active) {
-		rwlock_init(&gpc->lock);
-		mutex_init(&gpc->refresh_lock);
-
 		gpc->khva = NULL;
 		gpc->pfn = KVM_PFN_ERR_FAULT;
 		gpc->uhva = KVM_HVA_ERR_BAD;
 		gpc->vcpu = vcpu;
 		gpc->usage = usage;
 		gpc->valid = false;
-		gpc->active = true;
 
 		spin_lock(&kvm->gpc_lock);
 		list_add(&gpc->list, &kvm->gpc_list);
 		spin_unlock(&kvm->gpc_lock);
+
+		/*
+		 * Activate the cache after adding it to the list, a concurrent
+		 * refresh must not establish a mapping until the cache is
+		 * reachable by mmu_notifier events.
+		 */
+		write_lock_irq(&gpc->lock);
+		gpc->active = true;
+		write_unlock_irq(&gpc->lock);
 	}
 	return kvm_gfn_to_pfn_cache_refresh(kvm, gpc, gpa, len);
 }
-EXPORT_SYMBOL_GPL(kvm_gfn_to_pfn_cache_init);
+EXPORT_SYMBOL_GPL(kvm_gpc_activate);
 
-void kvm_gfn_to_pfn_cache_destroy(struct kvm *kvm, struct gfn_to_pfn_cache *gpc)
+void kvm_gpc_deactivate(struct kvm *kvm, struct gfn_to_pfn_cache *gpc)
 {
 	if (gpc->active) {
+		/*
+		 * Deactivate the cache before removing it from the list, KVM
+		 * must stall mmu_notifier events until all users go away, i.e.
+		 * until gpc->lock is dropped and refresh is guaranteed to fail.
+		 */
+		write_lock_irq(&gpc->lock);
+		gpc->active = false;
+		write_unlock_irq(&gpc->lock);
+
 		spin_lock(&kvm->gpc_lock);
 		list_del(&gpc->list);
 		spin_unlock(&kvm->gpc_lock);
 
 		kvm_gfn_to_pfn_cache_unmap(kvm, gpc);
-		gpc->active = false;
 	}
 }
-EXPORT_SYMBOL_GPL(kvm_gfn_to_pfn_cache_destroy);
+EXPORT_SYMBOL_GPL(kvm_gpc_deactivate);