get_next_shape_internal: Skip VM lock for single child case

If the shape has only one child, we check it lock-free without
compromising thread safety.

I haven't computed hard data as to how often that it the case,
but we can assume that it's not too rare for shapes to have
a single child that is often requested, typically when freezing
and object.

Author: byroot

shape.c

@@ -499,13 +499,23 @@ get_next_shape_internal(rb_shape_t * shape, ID id, enum shape_type shape_type, b
 
     *variation_created = false;
 
+    // Fast path: if the shape has a single child, we can check it without a lock
+    struct rb_id_table *edges = shape->edges;
+    if (edges && SINGLE_CHILD_P(edges)) {
+        rb_shape_t *child = SINGLE_CHILD(edges);
+        if (child->edge_name == id) {
+            return child;
+        }
+    }
+
     RB_VM_LOCK_ENTER();
     {
         // If the current shape has children
         if (shape->edges) {
             // Check if it only has one child
             if (SINGLE_CHILD_P(shape->edges)) {
-                rb_shape_t * child = SINGLE_CHILD(shape->edges);
+                rb_shape_t *child = SINGLE_CHILD(shape->edges);
+
                 // If the one child has a matching edge name, then great,
                 // we found what we want.
                 if (child->edge_name == id) {