KVM: x86: Support for disabling quirks

Introducing KVM_CAP_DISABLE_QUIRKS for disabling x86 quirks that were previous
created in order to overcome QEMU issues. Those issue were mostly result of
invalid VM BIOS.  Currently there are two quirks that can be disabled:

1. KVM_QUIRK_LINT0_REENABLED - LINT0 was enabled after boot
2. KVM_QUIRK_CD_NW_CLEARED - CD and NW are cleared after boot

These two issues are already resolved in recent releases of QEMU, and would
therefore be disabled by QEMU.
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Message-Id: <1428879221-29996-1-git-send-email-namit@cs.technion.ac.il>
[Report capability from KVM_CHECK_EXTENSION too. - Paolo]
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

Documentation/virtual/kvm/api.txt

@@ -959,7 +959,8 @@ documentation when it pops into existence).
 4.37 KVM_ENABLE_CAP
 
 Capability: KVM_CAP_ENABLE_CAP, KVM_CAP_ENABLE_CAP_VM
-Architectures: ppc, s390
+Architectures: x86 (only KVM_CAP_ENABLE_CAP_VM),
+	       mips (only KVM_CAP_ENABLE_CAP), ppc, s390
 Type: vcpu ioctl, vm ioctl (with KVM_CAP_ENABLE_CAP_VM)
 Parameters: struct kvm_enable_cap (in)
 Returns: 0 on success; -1 on error

arch/x86/include/asm/kvm_host.h

@@ -635,6 +635,8 @@ struct kvm_arch {
 	#endif
 
 	bool boot_vcpu_runs_old_kvmclock;
+
+	u64 disabled_quirks;
 };
 
 struct kvm_vm_stat {

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

@@ -345,4 +345,7 @@ struct kvm_xcrs {
 struct kvm_sync_regs {
 };
 
+#define KVM_QUIRK_LINT0_REENABLED	(1 << 0)
+#define KVM_QUIRK_CD_NW_CLEARED		(1 << 1)
+
 #endif /* _ASM_X86_KVM_H */

arch/x86/kvm/lapic.c

@@ -1577,6 +1577,7 @@ void kvm_lapic_reset(struct kvm_vcpu *vcpu)
 	for (i = 0; i < APIC_LVT_NUM; i++)
 		apic_set_reg(apic, APIC_LVTT + 0x10 * i, APIC_LVT_MASKED);
 	apic->lapic_timer.timer_mode = 0;
+	if (!(vcpu->kvm->arch.disabled_quirks & KVM_QUIRK_LINT0_REENABLED))
 		apic_set_reg(apic, APIC_LVT0,
 			     SET_APIC_DELIVERY_MODE(0, APIC_MODE_EXTINT));
 

arch/x86/kvm/svm.c

@@ -1575,6 +1575,7 @@ static void svm_set_cr0(struct kvm_vcpu *vcpu, unsigned long cr0)
 	 * does not do it - this results in some delay at
 	 * reboot
 	 */
+	if (!(vcpu->kvm->arch.disabled_quirks & KVM_QUIRK_CD_NW_CLEARED))
 		cr0 &= ~(X86_CR0_CD | X86_CR0_NW);
 	svm->vmcb->save.cr0 = cr0;
 	mark_dirty(svm->vmcb, VMCB_CR);

arch/x86/kvm/x86.c

@@ -2800,6 +2800,8 @@ int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext)
 	case KVM_CAP_HYPERV_TIME:
 	case KVM_CAP_IOAPIC_POLARITY_IGNORED:
 	case KVM_CAP_TSC_DEADLINE_TIMER:
+	case KVM_CAP_ENABLE_CAP_VM:
+	case KVM_CAP_DISABLE_QUIRKS:
 #ifdef CONFIG_KVM_DEVICE_ASSIGNMENT
 	case KVM_CAP_ASSIGN_DEV_IRQ:
 	case KVM_CAP_PCI_2_3:
@@ -3847,6 +3849,26 @@ int kvm_vm_ioctl_irq_line(struct kvm *kvm, struct kvm_irq_level *irq_event,
 	return 0;
 }
 
+static int kvm_vm_ioctl_enable_cap(struct kvm *kvm,
+				   struct kvm_enable_cap *cap)
+{
+	int r;
+
+	if (cap->flags)
+		return -EINVAL;
+
+	switch (cap->cap) {
+	case KVM_CAP_DISABLE_QUIRKS:
+		kvm->arch.disabled_quirks = cap->args[0];
+		r = 0;
+		break;
+	default:
+		r = -EINVAL;
+		break;
+	}
+	return r;
+}
+
 long kvm_arch_vm_ioctl(struct file *filp,
 		       unsigned int ioctl, unsigned long arg)
 {
@@ -4099,7 +4121,15 @@ long kvm_arch_vm_ioctl(struct file *filp,
 		r = 0;
 		break;
 	}
+	case KVM_ENABLE_CAP: {
+		struct kvm_enable_cap cap;
 
+		r = -EFAULT;
+		if (copy_from_user(&cap, argp, sizeof(cap)))
+			goto out;
+		r = kvm_vm_ioctl_enable_cap(kvm, &cap);
+		break;
+	}
 	default:
 		r = kvm_vm_ioctl_assigned_device(kvm, ioctl, arg);
 	}

include/uapi/linux/kvm.h

@@ -814,6 +814,7 @@ struct kvm_ppc_smmu_info {
 #define KVM_CAP_S390_INJECT_IRQ 113
 #define KVM_CAP_S390_IRQ_STATE 114
 #define KVM_CAP_PPC_HWRNG 115
+#define KVM_CAP_DISABLE_QUIRKS 116
 
 #ifdef KVM_CAP_IRQ_ROUTING