test(utils): 添加事件驱动模型测试

- 在 sqrt_test.go 中添加了 fastSqr1 测试函数，用于测试事件驱动模型
- 新增了 Event 和 Uint32AsyncEvent 类型用于测试
- 更新了 go.work、go.mod 和

common/utils/event/bench_test.go

@@ -0,0 +1,271 @@
+package bus_test
+
+import (
+	"sync/atomic"
+	"testing"
+
+	"github.com/badu/bus"
+)
+
+type Uint32SyncEvent struct {
+	u uint32
+}
+
+type Uint32AsyncEvent struct {
+	u uint32
+}
+
+func BenchmarkBroadcast_0008Sync(b *testing.B) {
+	topic := bus.NewTopic[Uint32SyncEvent]()
+	c := uint32(0)
+	for i := 0; i < 8; i++ {
+		topic.Sub(func(v Uint32SyncEvent) {
+			atomic.AddUint32(&c, v.u)
+		})
+	}
+
+	b.RunParallel(func(p *testing.PB) {
+		for p.Next() {
+			topic.Pub(Uint32SyncEvent{u: 1})
+		}
+	})
+}
+
+func BenchmarkBroadcast_0008Async(b *testing.B) {
+	topic := bus.NewTopic[Uint32AsyncEvent]()
+	c := uint32(0)
+	for i := 0; i < 8; i++ {
+		topic.Sub(func(v Uint32AsyncEvent) {
+			atomic.AddUint32(&c, v.u)
+		})
+	}
+
+	b.RunParallel(func(p *testing.PB) {
+		for p.Next() {
+			topic.PubAsync(Uint32AsyncEvent{u: 1})
+		}
+	})
+}
+
+func BenchmarkBroadcast_0008PtrSync(b *testing.B) {
+	topic := bus.NewTopic[*Uint32SyncEvent]()
+	c := uint32(0)
+	for i := 0; i < 8; i++ {
+		topic.Sub(func(v *Uint32SyncEvent) {
+			atomic.AddUint32(&c, v.u)
+		})
+	}
+
+	b.RunParallel(func(p *testing.PB) {
+		for p.Next() {
+			topic.Pub(&Uint32SyncEvent{u: 1})
+		}
+	})
+}
+
+func BenchmarkBroadcast_0008PtrAsync(b *testing.B) {
+	topic := bus.NewTopic[*Uint32AsyncEvent]()
+	c := uint32(0)
+	for i := 0; i < 8; i++ {
+		topic.Sub(func(v *Uint32AsyncEvent) {
+			atomic.AddUint32(&c, v.u)
+		})
+	}
+
+	b.RunParallel(func(p *testing.PB) {
+		for p.Next() {
+			topic.PubAsync(&Uint32AsyncEvent{u: 1})
+		}
+	})
+}
+
+func BenchmarkBroadcast_0256Sync(b *testing.B) {
+	topic := bus.NewTopic[Uint32SyncEvent]()
+	var c uint32
+	for i := 0; i < 256; i++ {
+		topic.Sub(func(v Uint32SyncEvent) {
+			atomic.AddUint32(&c, v.u)
+		})
+	}
+
+	b.RunParallel(func(p *testing.PB) {
+		for p.Next() {
+			topic.Pub(Uint32SyncEvent{u: 1})
+		}
+	})
+}
+
+func BenchmarkBroadcast_0256Async(b *testing.B) {
+	topic := bus.NewTopic[Uint32AsyncEvent]()
+	var c uint32
+	for i := 0; i < 256; i++ {
+		topic.Sub(func(v Uint32AsyncEvent) {
+			atomic.AddUint32(&c, v.u)
+		})
+	}
+
+	b.RunParallel(func(p *testing.PB) {
+		for p.Next() {
+			topic.PubAsync(Uint32AsyncEvent{u: 1})
+		}
+	})
+}
+
+func BenchmarkBroadcast_0256PtrSync(b *testing.B) {
+	topic := bus.NewTopic[*Uint32SyncEvent]()
+	var c uint32
+	for i := 0; i < 256; i++ {
+		topic.Sub(func(v *Uint32SyncEvent) {
+			atomic.AddUint32(&c, v.u)
+		})
+	}
+
+	b.RunParallel(func(p *testing.PB) {
+		for p.Next() {
+			topic.Pub(&Uint32SyncEvent{u: 1})
+		}
+	})
+}
+
+func BenchmarkBroadcast_0256PtrAsync(b *testing.B) {
+	topic := bus.NewTopic[*Uint32AsyncEvent]()
+	var c uint32
+	for i := 0; i < 256; i++ {
+		topic.Sub(func(v *Uint32AsyncEvent) {
+			atomic.AddUint32(&c, v.u)
+		})
+	}
+
+	b.RunParallel(func(p *testing.PB) {
+		for p.Next() {
+			topic.PubAsync(&Uint32AsyncEvent{u: 1})
+		}
+	})
+}
+
+func BenchmarkBroadcast_1kSync(b *testing.B) {
+	topic := bus.NewTopic[Uint32SyncEvent]()
+	var c uint32
+	for i := 0; i < 1024; i++ {
+		topic.Sub(func(v Uint32SyncEvent) {
+			atomic.AddUint32(&c, v.u)
+		})
+	}
+
+	b.RunParallel(func(p *testing.PB) {
+		for p.Next() {
+			topic.Pub(Uint32SyncEvent{u: 1})
+		}
+	})
+}
+
+func BenchmarkBroadcast_1kAsync(b *testing.B) {
+	topic := bus.NewTopic[Uint32AsyncEvent]()
+	var c uint32
+	for i := 0; i < 1024; i++ {
+		topic.Sub(func(v Uint32AsyncEvent) {
+			atomic.AddUint32(&c, v.u)
+		})
+	}
+
+	b.RunParallel(func(p *testing.PB) {
+		for p.Next() {
+			topic.PubAsync(Uint32AsyncEvent{u: 1})
+		}
+	})
+}
+
+func BenchmarkBroadcast_1kPtrSync(b *testing.B) {
+	topic := bus.NewTopic[*Uint32SyncEvent]()
+	var c uint32
+	for i := 0; i < 1024; i++ {
+		topic.Sub(func(v *Uint32SyncEvent) {
+			atomic.AddUint32(&c, v.u)
+		})
+	}
+
+	b.RunParallel(func(p *testing.PB) {
+		for p.Next() {
+			topic.Pub(&Uint32SyncEvent{u: 1})
+		}
+	})
+}
+
+func BenchmarkBroadcast_1kPtrAsync(b *testing.B) {
+	topic := bus.NewTopic[*Uint32AsyncEvent]()
+	var c uint32
+	for i := 0; i < 1024; i++ {
+		topic.Sub(func(v *Uint32AsyncEvent) {
+			atomic.AddUint32(&c, v.u)
+		})
+	}
+
+	b.RunParallel(func(p *testing.PB) {
+		for p.Next() {
+			topic.PubAsync(&Uint32AsyncEvent{u: 1})
+		}
+	})
+}
+func BenchmarkBroadcast_2kSync(b *testing.B) {
+	topic := bus.NewTopic[Uint32SyncEvent]()
+	var c uint32
+	for i := 0; i < 2048; i++ {
+		topic.Sub(func(v Uint32SyncEvent) {
+			atomic.AddUint32(&c, v.u)
+		})
+	}
+
+	b.RunParallel(func(p *testing.PB) {
+		for p.Next() {
+			topic.Pub(Uint32SyncEvent{u: 1})
+		}
+	})
+}
+
+func BenchmarkBroadcast_2kAsync(b *testing.B) {
+	topic := bus.NewTopic[Uint32AsyncEvent]()
+	var c uint32
+	for i := 0; i < 2048; i++ {
+		topic.Sub(func(v Uint32AsyncEvent) {
+			atomic.AddUint32(&c, v.u)
+		})
+	}
+
+	b.RunParallel(func(p *testing.PB) {
+		for p.Next() {
+			topic.PubAsync(Uint32AsyncEvent{u: 1})
+		}
+	})
+}
+
+func BenchmarkBroadcast_2kPtrSync(b *testing.B) {
+	topic := bus.NewTopic[*Uint32SyncEvent]()
+	var c uint32
+	for i := 0; i < 2048; i++ {
+		topic.Sub(func(v *Uint32SyncEvent) {
+			atomic.AddUint32(&c, v.u)
+		})
+	}
+
+	b.RunParallel(func(p *testing.PB) {
+		for p.Next() {
+			topic.Pub(&Uint32SyncEvent{u: 1})
+		}
+	})
+}
+
+func BenchmarkBroadcast_2kPtrAsync(b *testing.B) {
+	topic := bus.NewTopic[*Uint32AsyncEvent]()
+	var c uint32
+	for i := 0; i < 2048; i++ {
+		topic.Sub(func(v *Uint32AsyncEvent) {
+			atomic.AddUint32(&c, v.u)
+		})
+	}
+
+	b.RunParallel(func(p *testing.PB) {
+		for p.Next() {
+			topic.PubAsync(&Uint32AsyncEvent{u: 1})
+		}
+	})
+}