[PATCH v5 2/2] arm64: kvm: Introduce MTE VCPU feature

[Steven Price <steven.price@arm.com>]

Add a new VM feature 'KVM_ARM_CAP_MTE' which enables memory tagging
for a VM. This exposes the feature to the guest and requires the VMM to
have set PROT_MTE on all mappings exposed to the guest. This ensures
that the guest cannot see stale tags, and that the tags are correctly
saved/restored across swap.

Signed-off-by: Steven Price <steven.price@arm.com>
---
 arch/arm64/include/asm/kvm_emulate.h | 3 +++
 arch/arm64/include/asm/kvm_host.h    | 4 ++++
 arch/arm64/kvm/arm.c                 | 9 +++++++++
 arch/arm64/kvm/mmu.c                 | 6 ++++++
 arch/arm64/kvm/sys_regs.c            | 6 +++++-
 include/uapi/linux/kvm.h             | 1 +
 6 files changed, 28 insertions(+), 1 deletion(-)

diff --git a/arch/arm64/include/asm/kvm_emulate.h b/arch/arm64/include/asm/kvm_emulate.h
index 5ef2669ccd6c..7791ef044b7f 100644
--- a/arch/arm64/include/asm/kvm_emulate.h
+++ b/arch/arm64/include/asm/kvm_emulate.h
@@ -79,6 +79,9 @@ static inline void vcpu_reset_hcr(struct kvm_vcpu *vcpu)
 	if (cpus_have_const_cap(ARM64_MISMATCHED_CACHE_TYPE) ||
 	    vcpu_el1_is_32bit(vcpu))
 		vcpu->arch.hcr_el2 |= HCR_TID2;
+
+	if (kvm_has_mte(vcpu->kvm))
+		vcpu->arch.hcr_el2 |= HCR_ATA;
 }
 
 static inline unsigned long *vcpu_hcr(struct kvm_vcpu *vcpu)
diff --git a/arch/arm64/include/asm/kvm_host.h b/arch/arm64/include/asm/kvm_host.h
index d3e136343468..aeff10bc5b31 100644
--- a/arch/arm64/include/asm/kvm_host.h
+++ b/arch/arm64/include/asm/kvm_host.h
@@ -120,6 +120,8 @@ struct kvm_arch {
 	unsigned int pmuver;
 
 	u8 pfr0_csv2;
+	/* Memory Tagging Extension enabled for the guest */
+	bool mte_enabled;
 };
 
 struct kvm_vcpu_fault_info {
@@ -658,4 +660,6 @@ bool kvm_arm_vcpu_is_finalized(struct kvm_vcpu *vcpu);
 #define kvm_arm_vcpu_sve_finalized(vcpu) \
 	((vcpu)->arch.flags & KVM_ARM64_VCPU_SVE_FINALIZED)
 
+#define kvm_has_mte(kvm) (system_supports_mte() && (kvm)->arch.mte_enabled)
+
 #endif /* __ARM64_KVM_HOST_H__ */
diff --git a/arch/arm64/kvm/arm.c b/arch/arm64/kvm/arm.c
index c0ffb019ca8b..da4aeba1855c 100644
--- a/arch/arm64/kvm/arm.c
+++ b/arch/arm64/kvm/arm.c
@@ -89,6 +89,12 @@ int kvm_vm_ioctl_enable_cap(struct kvm *kvm,
 		r = 0;
 		kvm->arch.return_nisv_io_abort_to_user = true;
 		break;
+	case KVM_CAP_ARM_MTE:
+		if (!system_supports_mte() || kvm->created_vcpus)
+			return -EINVAL;
+		r = 0;
+		kvm->arch.mte_enabled = true;
+		break;
 	default:
 		r = -EINVAL;
 		break;
@@ -226,6 +232,9 @@ int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext)
 		 */
 		r = 1;
 		break;
+	case KVM_CAP_ARM_MTE:
+		r = system_supports_mte();
+		break;
 	case KVM_CAP_STEAL_TIME:
 		r = kvm_arm_pvtime_supported();
 		break;
diff --git a/arch/arm64/kvm/mmu.c b/arch/arm64/kvm/mmu.c
index 1a01da9fdc99..f804d2109b8c 100644
--- a/arch/arm64/kvm/mmu.c
+++ b/arch/arm64/kvm/mmu.c
@@ -815,6 +815,12 @@ static int user_mem_abort(struct kvm_vcpu *vcpu, phys_addr_t fault_ipa,
 	if (vma_pagesize == PMD_SIZE || vma_pagesize == PUD_SIZE)
 		fault_ipa &= ~(vma_pagesize - 1);
 
+	/* VMA regions must be mapped with PROT_MTE when VM has MTE enabled */
+	if (kvm_has_mte(kvm) && !(vma->vm_flags & VM_MTE)) {
+		pr_err_ratelimited("Page not mapped VM_MTE in MTE guest\n");
+		return -EFAULT;
+	}
+
 	gfn = fault_ipa >> PAGE_SHIFT;
 	mmap_read_unlock(current->mm);
 
diff --git a/arch/arm64/kvm/sys_regs.c b/arch/arm64/kvm/sys_regs.c
index 4792d5249f07..469b0ef3eb07 100644
--- a/arch/arm64/kvm/sys_regs.c
+++ b/arch/arm64/kvm/sys_regs.c
@@ -1123,7 +1123,8 @@ static u64 read_id_reg(const struct kvm_vcpu *vcpu,
 		val &= ~(0xfUL << ID_AA64PFR0_CSV2_SHIFT);
 		val |= ((u64)vcpu->kvm->arch.pfr0_csv2 << ID_AA64PFR0_CSV2_SHIFT);
 	} else if (id == SYS_ID_AA64PFR1_EL1) {
-		val &= ~(0xfUL << ID_AA64PFR1_MTE_SHIFT);
+		if (!kvm_has_mte(vcpu->kvm))
+			val &= ~(0xfUL << ID_AA64PFR1_MTE_SHIFT);
 	} else if (id == SYS_ID_AA64ISAR1_EL1 && !vcpu_has_ptrauth(vcpu)) {
 		val &= ~((0xfUL << ID_AA64ISAR1_APA_SHIFT) |
 			 (0xfUL << ID_AA64ISAR1_API_SHIFT) |
@@ -1369,6 +1370,9 @@ static bool access_ccsidr(struct kvm_vcpu *vcpu, struct sys_reg_params *p,
 static unsigned int mte_visibility(const struct kvm_vcpu *vcpu,
 				   const struct sys_reg_desc *rd)
 {
+	if (kvm_has_mte(vcpu->kvm))
+		return 0;
+
 	return REG_HIDDEN;
 }
 
diff --git a/include/uapi/linux/kvm.h b/include/uapi/linux/kvm.h
index ca41220b40b8..3e6fb5b580a9 100644
--- a/include/uapi/linux/kvm.h
+++ b/include/uapi/linux/kvm.h
@@ -1053,6 +1053,7 @@ struct kvm_ppc_resize_hpt {
 #define KVM_CAP_X86_USER_SPACE_MSR 188
 #define KVM_CAP_X86_MSR_FILTER 189
 #define KVM_CAP_ENFORCE_PV_FEATURE_CPUID 190
+#define KVM_CAP_ARM_MTE 191
 
 #ifdef KVM_CAP_IRQ_ROUTING
 
-- 
2.20.1