KVM: x86: Add KVM_RUN_X86_GUEST_MODE kvm_run flag

When a vCPU is interrupted by a signal while running a nested guest,
KVM will exit to userspace with L2 state. However, userspace has no
way to know whether it sees L1 or L2 state (besides calling
KVM_GET_STATS_FD, which does not have a stable ABI).

This causes multiple problems:

The simplest one is L2 state corruption when userspace marks the sregs
as dirty. See this mailing list thread [1] for a complete discussion.

Another problem is that if userspace decides to continue by emulating
instructions, it will unknowingly emulate with L2 state as if L1
doesn't exist, which can be considered a weird guest escape.

Introduce a new flag, KVM_RUN_X86_GUEST_MODE, in the kvm_run data
structure, which is set when the vCPU exited while running a nested
guest.  Also introduce a new capability, KVM_CAP_X86_GUEST_MODE, to
advertise the functionality to userspace.

[1] https://lore.kernel.org/kvm/[email protected]/T/#m280aadcb2e10ae02c191a7dc4ed4b711a74b1f55

Signed-off-by: Thomas Prescher <[email protected]>
Signed-off-by: Julian Stecklina <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Sean Christopherson <[email protected]>

Author: tpressure

Documentation/virt/kvm/api.rst

@@ -6419,6 +6419,9 @@ affect the device's behavior. Current defined flags::
   #define KVM_RUN_X86_SMM          (1 << 0)
   /* x86, set if bus lock detected in VM */
   #define KVM_RUN_X86_BUS_LOCK     (1 << 1)
+  /* x86, set if the VCPU is executing a nested (L2) guest */
+  #define KVM_RUN_X86_GUEST_MODE   (1 << 2)
+
   /* arm64, set for KVM_EXIT_DEBUG */
   #define KVM_DEBUG_ARCH_HSR_HIGH_VALID  (1 << 0)
 
@@ -8089,6 +8092,20 @@ by KVM_CHECK_EXTENSION.
 Note: Userspace is responsible for correctly configuring CPUID 0x15, a.k.a. the
 core crystal clock frequency, if a non-zero CPUID 0x15 is exposed to the guest.
 
+7.36 KVM_CAP_X86_GUEST_MODE
+------------------------------
+
+:Architectures: x86
+:Returns: Informational only, -EINVAL on direct KVM_ENABLE_CAP.
+
+The presence of this capability indicates that KVM_RUN will update the
+KVM_RUN_X86_GUEST_MODE bit in kvm_run.flags to indicate whether the
+vCPU was executing nested guest code when it exited.
+
+KVM exits with the register state of either the L1 or L2 guest
+depending on which executed at the time of an exit. Userspace must
+take care to differentiate between these cases.
+
 8. Other capabilities.
 ======================
 

arch/x86/include/uapi/asm/kvm.h

@@ -106,6 +106,7 @@ struct kvm_ioapic_state {
 
 #define KVM_RUN_X86_SMM		 (1 << 0)
 #define KVM_RUN_X86_BUS_LOCK     (1 << 1)
+#define KVM_RUN_X86_GUEST_MODE   (1 << 2)
 
 /* for KVM_GET_REGS and KVM_SET_REGS */
 struct kvm_regs {

arch/x86/kvm/x86.c

@@ -4704,6 +4704,7 @@ int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext)
 	case KVM_CAP_VM_DISABLE_NX_HUGE_PAGES:
 	case KVM_CAP_IRQFD_RESAMPLE:
 	case KVM_CAP_MEMORY_FAULT_INFO:
+	case KVM_CAP_X86_GUEST_MODE:
 		r = 1;
 		break;
 	case KVM_CAP_X86_APIC_BUS_CYCLES_NS:
@@ -10277,6 +10278,8 @@ static void post_kvm_run_save(struct kvm_vcpu *vcpu)
 
 	if (is_smm(vcpu))
 		kvm_run->flags |= KVM_RUN_X86_SMM;
+	if (is_guest_mode(vcpu))
+		kvm_run->flags |= KVM_RUN_X86_GUEST_MODE;
 }
 
 static void update_cr8_intercept(struct kvm_vcpu *vcpu)

include/uapi/linux/kvm.h

@@ -918,6 +918,7 @@ struct kvm_enable_cap {
 #define KVM_CAP_GUEST_MEMFD 234
 #define KVM_CAP_VM_TYPES 235
 #define KVM_CAP_X86_APIC_BUS_CYCLES_NS 236
+#define KVM_CAP_X86_GUEST_MODE 237
 
 struct kvm_irq_routing_irqchip {
 	__u32 irqchip;