KVM: arm64: Create each pKVM hyp vcpu after its corresponding host vcpu

Instead of creating and initializing _all_ hyp vcpus in pKVM when
the first host vcpu runs for the first time, initialize _each_
hyp vcpu in conjunction with its corresponding host vcpu.

Some of the host vcpu state (e.g., system registers and traps
values) is not initialized until the first time the host vcpu is
run. Therefore, initializing a hyp vcpu before its corresponding
host vcpu has run for the first time might not view the complete
host state of these vcpus.

Additionally, this behavior is inline with non-protected modes.

Acked-by: Will Deacon <will@kernel.org>
Reviewed-by: Marc Zyngier <maz@kernel.org>
Signed-off-by: Fuad Tabba <tabba@google.com>
Link: https://lore.kernel.org/r/20250314111832.4137161-5-tabba@google.com
Signed-off-by: Oliver Upton <oliver.upton@linux.dev>

Author: Fuad Tabba

arch/arm64/include/asm/kvm_host.h

@@ -873,6 +873,8 @@ struct kvm_vcpu_arch {
 #define VCPU_INITIALIZED	__vcpu_single_flag(cflags, BIT(0))
 /* SVE config completed */
 #define VCPU_SVE_FINALIZED	__vcpu_single_flag(cflags, BIT(1))
+/* pKVM VCPU setup completed */
+#define VCPU_PKVM_FINALIZED	__vcpu_single_flag(cflags, BIT(2))
 
 /* Exception pending */
 #define PENDING_EXCEPTION	__vcpu_single_flag(iflags, BIT(0))

arch/arm64/include/asm/kvm_pkvm.h

@@ -19,6 +19,7 @@
 int pkvm_init_host_vm(struct kvm *kvm);
 int pkvm_create_hyp_vm(struct kvm *kvm);
 void pkvm_destroy_hyp_vm(struct kvm *kvm);
+int pkvm_create_hyp_vcpu(struct kvm_vcpu *vcpu);
 
 /*
  * This functions as an allow-list of protected VM capabilities.

arch/arm64/kvm/arm.c

@@ -833,6 +833,10 @@ int kvm_arch_vcpu_run_pid_change(struct kvm_vcpu *vcpu)
 		ret = pkvm_create_hyp_vm(kvm);
 		if (ret)
 			return ret;
+
+		ret = pkvm_create_hyp_vcpu(vcpu);
+		if (ret)
+			return ret;
 	}
 
 	mutex_lock(&kvm->arch.config_lock);

arch/arm64/kvm/hyp/include/nvhe/pkvm.h

@@ -43,12 +43,6 @@ struct pkvm_hyp_vm {
 	struct hyp_pool pool;
 	hyp_spinlock_t lock;
 
-	/*
-	 * The number of vcpus initialized and ready to run.
-	 * Modifying this is protected by 'vm_table_lock'.
-	 */
-	unsigned int nr_vcpus;
-
 	/* Array of the hyp vCPU structures for this VM. */
 	struct pkvm_hyp_vcpu *vcpus[];
 };

arch/arm64/kvm/hyp/nvhe/pkvm.c

@@ -245,10 +245,12 @@ struct pkvm_hyp_vcpu *pkvm_load_hyp_vcpu(pkvm_handle_t handle,
 
 	hyp_spin_lock(&vm_table_lock);
 	hyp_vm = get_vm_by_handle(handle);
-	if (!hyp_vm || hyp_vm->nr_vcpus <= vcpu_idx)
+	if (!hyp_vm || hyp_vm->kvm.created_vcpus <= vcpu_idx)
 		goto unlock;
 
 	hyp_vcpu = hyp_vm->vcpus[vcpu_idx];
+	if (!hyp_vcpu)
+		goto unlock;
 
 	/* Ensure vcpu isn't loaded on more than one cpu simultaneously. */
 	if (unlikely(hyp_vcpu->loaded_hyp_vcpu)) {
@@ -367,8 +369,14 @@ static void unpin_host_vcpus(struct pkvm_hyp_vcpu *hyp_vcpus[],
 {
 	int i;
 
-	for (i = 0; i < nr_vcpus; i++)
-		unpin_host_vcpu(hyp_vcpus[i]->host_vcpu);
+	for (i = 0; i < nr_vcpus; i++) {
+		struct pkvm_hyp_vcpu *hyp_vcpu = hyp_vcpus[i];
+
+		if (!hyp_vcpu)
+			continue;
+
+		unpin_host_vcpu(hyp_vcpu->host_vcpu);
+	}
 }
 
 static void init_pkvm_hyp_vm(struct kvm *host_kvm, struct pkvm_hyp_vm *hyp_vm,
@@ -392,24 +400,18 @@ static void pkvm_vcpu_init_sve(struct pkvm_hyp_vcpu *hyp_vcpu, struct kvm_vcpu *
 
 static int init_pkvm_hyp_vcpu(struct pkvm_hyp_vcpu *hyp_vcpu,
 			      struct pkvm_hyp_vm *hyp_vm,
-			      struct kvm_vcpu *host_vcpu,
-			      unsigned int vcpu_idx)
+			      struct kvm_vcpu *host_vcpu)
 {
 	int ret = 0;
 
 	if (hyp_pin_shared_mem(host_vcpu, host_vcpu + 1))
 		return -EBUSY;
 
-	if (host_vcpu->vcpu_idx != vcpu_idx) {
-		ret = -EINVAL;
-		goto done;
-	}
-
 	hyp_vcpu->host_vcpu = host_vcpu;
 
 	hyp_vcpu->vcpu.kvm = &hyp_vm->kvm;
 	hyp_vcpu->vcpu.vcpu_id = READ_ONCE(host_vcpu->vcpu_id);
-	hyp_vcpu->vcpu.vcpu_idx = vcpu_idx;
+	hyp_vcpu->vcpu.vcpu_idx = READ_ONCE(host_vcpu->vcpu_idx);
 
 	hyp_vcpu->vcpu.arch.hw_mmu = &hyp_vm->kvm.arch.mmu;
 	hyp_vcpu->vcpu.arch.cflags = READ_ONCE(host_vcpu->arch.cflags);
@@ -647,27 +649,28 @@ int __pkvm_init_vcpu(pkvm_handle_t handle, struct kvm_vcpu *host_vcpu,
 		goto unlock;
 	}
 
-	idx = hyp_vm->nr_vcpus;
+	ret = init_pkvm_hyp_vcpu(hyp_vcpu, hyp_vm, host_vcpu);
+	if (ret)
+		goto unlock;
+
+	idx = hyp_vcpu->vcpu.vcpu_idx;
 	if (idx >= hyp_vm->kvm.created_vcpus) {
 		ret = -EINVAL;
 		goto unlock;
 	}
 
-	ret = init_pkvm_hyp_vcpu(hyp_vcpu, hyp_vm, host_vcpu, idx);
-	if (ret)
+	if (hyp_vm->vcpus[idx]) {
+		ret = -EINVAL;
 		goto unlock;
+	}
 
 	hyp_vm->vcpus[idx] = hyp_vcpu;
-	hyp_vm->nr_vcpus++;
 unlock:
 	hyp_spin_unlock(&vm_table_lock);
 
-	if (ret) {
+	if (ret)
 		unmap_donated_memory(hyp_vcpu, sizeof(*hyp_vcpu));
-		return ret;
-	}
-
-	return 0;
+	return ret;
 }
 
 static void
@@ -714,11 +717,17 @@ int __pkvm_teardown_vm(pkvm_handle_t handle)
 	mc = &host_kvm->arch.pkvm.teardown_mc;
 	stage2_mc = &host_kvm->arch.pkvm.stage2_teardown_mc;
 	reclaim_pgtable_pages(hyp_vm, stage2_mc);
-	unpin_host_vcpus(hyp_vm->vcpus, hyp_vm->nr_vcpus);
+	unpin_host_vcpus(hyp_vm->vcpus, hyp_vm->kvm.created_vcpus);
 
-	for (idx = 0; idx < hyp_vm->nr_vcpus; ++idx) {
+	/* Push the metadata pages to the teardown memcache */
+	for (idx = 0; idx < hyp_vm->kvm.created_vcpus; ++idx) {
 		struct pkvm_hyp_vcpu *hyp_vcpu = hyp_vm->vcpus[idx];
-		struct kvm_hyp_memcache *vcpu_mc = &hyp_vcpu->vcpu.arch.pkvm_memcache;
+		struct kvm_hyp_memcache *vcpu_mc;
+
+		if (!hyp_vcpu)
+			continue;
+
+		vcpu_mc = &hyp_vcpu->vcpu.arch.pkvm_memcache;
 
 		while (vcpu_mc->nr_pages) {
 			void *addr = pop_hyp_memcache(vcpu_mc, hyp_phys_to_virt);

arch/arm64/kvm/pkvm.c

@@ -128,7 +128,9 @@ static int __pkvm_create_hyp_vcpu(struct kvm_vcpu *vcpu)
 		return -ENOMEM;
 
 	ret = kvm_call_hyp_nvhe(__pkvm_init_vcpu, handle, vcpu, hyp_vcpu);
-	if (ret)
+	if (!ret)
+		vcpu_set_flag(vcpu, VCPU_PKVM_FINALIZED);
+	else
 		free_pages_exact(hyp_vcpu, hyp_vcpu_sz);
 
 	return ret;
@@ -147,9 +149,7 @@ static int __pkvm_create_hyp_vcpu(struct kvm_vcpu *vcpu)
 static int __pkvm_create_hyp_vm(struct kvm *host_kvm)
 {
 	size_t pgd_sz, hyp_vm_sz;
-	struct kvm_vcpu *host_vcpu;
 	void *pgd, *hyp_vm;
-	unsigned long idx;
 	int ret;
 
 	if (host_kvm->created_vcpus < 1)
@@ -185,17 +185,7 @@ static int __pkvm_create_hyp_vm(struct kvm *host_kvm)
 	host_kvm->arch.pkvm.stage2_teardown_mc.flags |= HYP_MEMCACHE_ACCOUNT_STAGE2;
 	kvm_account_pgtable_pages(pgd, pgd_sz / PAGE_SIZE);
 
-	kvm_for_each_vcpu(idx, host_vcpu, host_kvm) {
-		ret = __pkvm_create_hyp_vcpu(host_vcpu);
-		if (ret)
-			goto destroy_vm;
-	}
-
 	return 0;
-
-destroy_vm:
-	__pkvm_destroy_hyp_vm(host_kvm);
-	return ret;
 free_vm:
 	free_pages_exact(hyp_vm, hyp_vm_sz);
 free_pgd:
@@ -215,6 +205,18 @@ int pkvm_create_hyp_vm(struct kvm *host_kvm)
 	return ret;
 }
 
+int pkvm_create_hyp_vcpu(struct kvm_vcpu *vcpu)
+{
+	int ret = 0;
+
+	mutex_lock(&vcpu->kvm->arch.config_lock);
+	if (!vcpu_get_flag(vcpu, VCPU_PKVM_FINALIZED))
+		ret = __pkvm_create_hyp_vcpu(vcpu);
+	mutex_unlock(&vcpu->kvm->arch.config_lock);
+
+	return ret;
+}
+
 void pkvm_destroy_hyp_vm(struct kvm *host_kvm)
 {
 	mutex_lock(&host_kvm->arch.config_lock);