[CareerCup] 16.4 A Lock Without Deadlocks 无死锁的锁
16.4 Design a class which provides a lock only if there are no possible deadlocks.
有很多方法可以避免死锁的发生,一个常用的方法是列出所需要的锁,然后判断锁上这些锁后会不会发生死锁,比如有如下的锁的顺序:
A = {1, 2, 3, 4}
B = {1, 3, 5}
C = {7, 5, 9, 2}
这是有可能产生死锁的,比如当A锁上2等待3,当B锁上3等待5,当C锁上5等待2,我们可以将其看做图,2连上3,3连上5,5连上2,那么就会有环。一条边(w,v)表示锁上v后马上锁w,那么上述里子的图中的边为(1, 2), (2, 3), (3, 4), (1, 3), (3, 5), (7, 5), (5, 9), (9, 2)。那么我们在检测是否有死锁的情况就是要找图中是否存在环,我们可以用DFS来搜索所有的相连的部分,我们用DFS需要标记点的状态,我们首先定义个LockFactory类,用来保存锁的序列,比如上面的A,B,C,然后对于每一个锁序列,我们先将所有的锁都链起来,然后标记为false,然后用DFS判断是否有环,如果有环,则断开当前锁序列之间的所有连接,参见代码如下;
import java.util.Hashtable;
import java.util.LinkedList;
import java.util.concurrent.locks.Lock; public class LockFactory {
private static LockFactory instance;
private int numberOfLocks = 5;
private LockNode[] locks;
private Hashtable<Integer, LinkedList<LockNode>> lockOrder; private LockFactory(int count) {
numberOfLocks = count;
locks = new LockNode[numberOfLocks];
lockOrder = new Hashtable<Integer, LinkedList<LockNode>>();
for (int i = 0; i < numberOfLocks; ++i) {
locks[i] = new LockNode(i, count);
}
} public static LockFactory getInstance() {
return instance;
} public static LockFactory initialize(int count) {
if (instance == null) {
instance = new LockFactory(count);
}
return instance;
} public boolean hasCycle(Hashtable<Integer, Boolean> touchedNodes, int[] resourcesInOrder) {
for (int resource : resourcesInOrder) {
if (touchedNodes.get(resource) == false) {
LockNode n = locks[resource];
if (n.hasCycle(touchedNodes)) {
return true;
}
}
}
return false;
} public boolean declare(int ownerId, int[] resourcesInOrder) {
Hashtable<Integer, Boolean> touchedNodes = new Hashtable<Integer, Boolean>();
int index = 1;
touchedNodes.put(resourcesInOrder[0], false);
for (index = 1; index < resourcesInOrder.length; ++index) {
LockNode pre = locks[resourcesInOrder[index - 1]];
LockNode cur = locks[resourcesInOrder[index]];
pre.joinTo(cur);
touchedNodes.put(resourcesInOrder[index], false);
}
if (hasCycle(touchedNodes, resourcesInOrder)) {
for (int j = 1; j < resourcesInOrder.length; ++j) {
LockNode p = locks[resourcesInOrder[j - 1]];
LockNode c = locks[resourcesInOrder[j]];
p.remove(c);
}
return false;
}
LinkedList<LockNode> list = new LinkedList<LockNode>();
for (int i = 0; i < resourcesInOrder.length; ++i) {
LockNode resource = locks[resourcesInOrder[i]];
list.add(resource);
}
lockOrder.put(ownerId, list);
return true;
} public Lock getLock(int ownerId, int resourceId) {
LinkedList<LockNode> list = lockOrder.get(ownerId);
if (list == null) return null;
LockNode head = list.getFirst();
if (head.getId() == resourceId) {
list.removeFirst();
return head.getLock();
}
return null;
}
} import java.util.ArrayList;
import java.util.Hashtable;
import java.util.concurrent.locks.Lock;
import java.util.concurrent.locks.ReentrantLock; public class LockNode {
public enum VisitState {
FRESH, VISITING, VISITED
};
private ArrayList<LockNode> children;
private int lockId;
private Lock lock;
private int maxLocks; public LockNode(int id, int max) {
lockId = id;
children = new ArrayList<LockNode>();
maxLocks = max;
} public void joinTo(LockNode node) {
children.add(node);
} public void remove(LockNode node) {
children.remove(node);
} public boolean hasCycle(Hashtable<Integer, Boolean> touchedNodes) {
VisitState[] visited = new VisitState[maxLocks];
for (int i = 0; i < maxLocks; ++i) {
visited[i] = VisitState.FRESH;
}
return hasCycle(visited, touchedNodes);
} public boolean hasCycle(VisitState[] visited, Hashtable<Integer, Boolean> touchedNodes) {
if (touchedNodes.containsKey(lockId)) {
touchedNodes.put(lockId, true);
}
if (visited[lockId] == VisitState.VISITING) {
return true;
} else if (visited[lockId] == VisitState.FRESH) {
visited[lockId] = VisitState.VISITING;
for (LockNode n : children) {
if (n.hasCycle(visited, touchedNodes)) {
return true;
}
}
visited[lockId] = VisitState.VISITED;
}
return false;
} public Lock getLock() {
if (lock == null) {
lock = new ReentrantLock();
}
return lock;
} public int getId() {
return lockId;
}
} public class j {
public static void main(String[] args) {
int[] res1 = {1, 2, 3, 4};
int[] res2 = {1, 5, 4, 1};
int[] res3 = {1, 4, 5};
LockFactory.initialize(10); LockFactory lf = LockFactory.getInstance();
System.out.println(lf.declare(1, res1));
System.out.println(lf.declare(2, res2));
System.out.println(lf.declare(3, res3)); System.out.println(lf.getLock(1, 1));
System.out.println(lf.getLock(1, 2));
System.out.println(lf.getLock(1, 3));
System.out.println(lf.getLock(1, 4));
System.out.println(lf.getLock(2, 1));
System.out.println(lf.getLock(2, 5));
System.out.println(lf.getLock(2, 4));
System.out.println(lf.getLock(3, 1));
System.out.println(lf.getLock(3, 4));
System.out.println(lf.getLock(3, 5));
}
}
[CareerCup] 16.4 A Lock Without Deadlocks 无死锁的锁的更多相关文章
- 转:synchronized和LOCK的实现原理---深入JVM锁机制
JVM底层又是如何实现synchronized的? 目前在Java中存在两种锁机制:synchronized和Lock,Lock接口及其实现类是JDK5增加的内容,其作者是大名鼎鼎的并发专家Doug ...
- 015-线程同步-synchronized几种加锁方式、Java对象头和Monitor、Mutex Lock、JDK1.6对synchronized锁的优化实现
一.synchronized概述基本使用 为确保共享变量不会出现并发问题,通常会对修改共享变量的代码块用synchronized加锁,确保同一时刻只有一个线程在修改共享变量,从而避免并发问题. syn ...
- 16、Java并发性和多线程-死锁
以下内容转自http://ifeve.com/deadlock/: 死锁是两个或更多线程阻塞着等待其它处于死锁状态的线程所持有的锁.死锁通常发生在多个线程同时但以不同的顺序请求同一组锁的时候. 例如, ...
- Lock方法是用于数据库的锁机制,
Lock方法是用于数据库的锁机制,如果在查询或者执行操作的时候使用: lock(true); 复制代码 就会自动在生成的SQL语句最后加上 FOR UPDATE或者FOR UPDATE NOWAI ...
- [CareerCup] 16.3 Dining Philosophers 哲学家聚餐问题
16.3 In the famous dining philosophers problem, a bunch of philosophers are sitting around a circula ...
- [CareerCup] 16.5 Semphore 信号旗
16.5 Suppose we have the following code:public class Foo { public Foo() { . . . } public void first( ...
- [CareerCup] 16.6 Synchronized Method 同步方法
16.6 You are given a class with synchronized method A and a normal method B. If you have two threads ...
- [CareerCup] 16.2 Measure Time in a Context Switch 测量上下文转换的时间
16.2 How would you measure the time spent in a context switch? 上下文转换发生在两个进程之间,比如让一个等待进程进入执行和让一个运行进程进 ...
- [CareerCup] 16.1 Thread and Process 线程和进程
16.1 What's the difference between a thread and a process? 进程Process是程序执行时的一个实例.一个进程是被分配系统资源的独立单元,每个 ...
随机推荐
- 常见IE浏览器bug及其修复方案(双外边距、3像素偏移、绝对定位)
1. 双外边距浮动bug IE6和更低版本中存在双外边距浮动bug,顾名思义,这个Windows bug使任何浮动元素上的外边距加倍 bug重现: <!DOCTYPE html> < ...
- linux中socket的理解
对linux中socket的理解 一.socket 一般来说socket有一个别名也叫做套接字. socket起源于Unix,都可以用“打开open –> 读写write/read –> ...
- JMockit
[TOC] 简介 JMockit是基于JavaSE5中的java.lang.instrument包开发,内部使用ASM库来动态修改java的字节码,使得java这种静态语言可以想动态脚本语言一样动态设 ...
- HDU 4005 The war Tarjan+dp
The war Problem Description In the war, the intelligence about the enemy is very important. Now, o ...
- 孙鑫VC学习笔记:多线程编程
孙鑫VC学习笔记:多线程编程 SkySeraph Dec 11st 2010 HQU Email:zgzhaobo@gmail.com QQ:452728574 Latest Modified ...
- sql 根据指定条件获取一个字段批量获取数据插入另外一张表字段中+MD5加密
/****** Object: StoredProcedure [dbo].[getSplitValue] Script Date: 03/13/2014 13:58:12 ******/ SET A ...
- MIT 6.828 JOS学习笔记2. Lab 1 Part 1.2: PC bootstrap
Lab 1 Part 1: PC bootstrap 我们继续~ PC机的物理地址空间 这一节我们将深入的探究到底PC是如何启动的.首先我们看一下通常一个PC的物理地址空间是如何布局的: ...
- JQuery学习之语法
1.JQuery语法就是通过选取HTML元素,并对选取的元素执行某些操作: 基础语法:$(selector).action() (1)美元符号定义jQuery (2)选择符(selector)查询和查 ...
- CSS3-网站导航,transform,transition
网站导航: 1.a:link visited hover active的顺序是很重要的,如果改变顺序,则hover以及active的状态不起作用 2.<a href=" ...
- Codeforces Round #334 (Div. 2)
水 A - Uncowed Forces #include <bits/stdc++.h> using namespace std; typedef long long ll; const ...