commit a49ba165881de350dede17436e0965b7404d9b22
author Madhukar Pappireddy <madhukar.pappireddy@arm.com> Mon Nov 25 09:40:45 2024 -0600
committer Joao Alves <joao.alves@arm.com> Wed Jan 29 19:18:00 2025 +0100
tree a58a8d18aa34778a0ce1301d531744f6bd8f1f28
parent 587f9ba8575d1566e3108a68f8bef17bb303818c

refactor: maintain the boot order list with VMs

In the present implementation, the boot order is maintained through a
list of nodes contained by vCPU data structures.

Boot order protocol is a concept that naturally aligns with VMs. In
subsequent patches, we depend on the boot order for bootstrapping
execution contexts of MP endpoints on secondary CPUs. This motivates
us to maintain the boot order list with elements representing their
parent VM entries. This makes it simpler to obtain the vCPU to be
bootstrapped for initialization when a secondary CPU is turned on.

Change-Id: I333fa2b3188db99b2b52e7e43988a7a7b9fb7aef
Signed-off-by: Madhukar Pappireddy <madhukar.pappireddy@arm.com>

src/vm.c

diff --git a/src/vm.c b/src/vm.c
index 8e4930c..e439a49 100644
--- a/src/vm.c
+++ b/src/vm.c
@@ -28,6 +28,12 @@
 static ffa_vm_count_t vm_count;
 
 /**
+ * The `boot_list` is a special entry in the circular linked list maintained by
+ * the partition manager and serves as both the start and end of the list.
+ */
+static struct list_entry boot_list = LIST_INIT(boot_list);
+
+/**
  * Counters on the status of notifications in the system. It helps to improve
  * the information retrieved by the receiver scheduler.
  */
@@ -94,6 +100,7 @@
 	}
 
 	vm_notifications_init(vm, vcpu_count, ppool);
+	list_init(&vm->boot_list_node);
 	return vm;
 }
 
@@ -1148,3 +1155,68 @@
 
 	return int_desc;
 }
+
+/**
+ * The 'boot_list' is used as the start and end of the list.
+ * Start: the nodes it points to is the first VM to boot.
+ * End: the last node's next points to the entry.
+ */
+static bool vm_is_boot_list_end(struct vm *vm)
+{
+	return vm->boot_list_node.next == &boot_list;
+}
+
+/**
+ * Gets the first partition to boot, according to Boot Protocol from FF-A spec.
+ */
+struct vm *vm_get_boot_vm(void)
+{
+	assert(!list_empty(&boot_list));
+
+	return CONTAINER_OF(boot_list.next, struct vm, boot_list_node);
+}
+
+/**
+ * Returns the next element in the boot order list, if there is one.
+ */
+struct vm *vm_get_next_boot(struct vm *vm)
+{
+	return vm_is_boot_list_end(vm)
+		       ? NULL
+		       : CONTAINER_OF(vm->boot_list_node.next, struct vm,
+				      boot_list_node);
+}
+
+/**
+ * Insert in boot list, sorted by `boot_order` parameter in the vm structure
+ * and rooted in `first_boot_vm`.
+ */
+void vm_update_boot(struct vm *vm)
+{
+	struct vm *current_vm = NULL;
+
+	if (list_empty(&boot_list)) {
+		list_prepend(&boot_list, &vm->boot_list_node);
+		return;
+	}
+
+	/*
+	 * When getting to this point the first insertion should have
+	 * been done.
+	 */
+	current_vm = vm_get_boot_vm();
+	assert(current_vm != NULL);
+
+	/*
+	 * Iterate until the position is found according to boot order, or
+	 * until we reach end of the list.
+	 */
+	while (!vm_is_boot_list_end(current_vm) &&
+	       current_vm->boot_order <= vm->boot_order) {
+		current_vm = vm_get_next_boot(current_vm);
+	}
+
+	current_vm->boot_order > vm->boot_order
+		? list_prepend(&current_vm->boot_list_node, &vm->boot_list_node)
+		: list_append(&current_vm->boot_list_node, &vm->boot_list_node);
+}