我創建了一個程序，將線性搜索（搜索 -1）劃分為 4 個單獨的線程。public class main { static boolean found = false; public static void main(String[] args) { // TODO Auto-generated method stub int threadCount = 4; //amount of threads to use Random rand = new Random(); Searcher[] s_arr = new Searcher[threadCount]; //array of threads int[] arr = new int[10000]; //array to search through for (int i = 0; i < arr.length; i++) //randomizing #'s in array arr[i] = (int) (rand.nextFloat() * 1000); int randIndex = rand.nextInt(arr.length); //choose random index arr[randIndex] = -1; //set random index to = -1 for (int i = 0; i < threadCount; i++) { // s_arr[i] = new Searcher(Arrays.copyOfRange(arr, i * (arr.length/threadCount), (i+1) * (arr.length/threadCount)), (int) (i), i); //assign subarray for this thread to search through System.out.println(s_arr[i].wait); s_arr[i].start(); } //CODE IN QUESTION HERE ---------------------------- //while (!found) ; while (!found) //wait until value is found { try { Thread.sleep(1); } catch (InterruptedException e) { // TODO Auto-generated catch block e.printStackTrace(); } }我已經標記出了有問題的代碼，在使用第一個（注釋掉的）while 循環時，程序不會超出該點，但是如果我切換并使用它正下方的另一個 while 循環（強制它的那個循環）每個迭代器等待 1 毫秒）程序運行得很好。為什么會這樣？是否有更有效/實用的方法來完成這項任務？
1 回答
不負相思意
TA貢獻1777條經驗 獲得超10個贊
在空循環語句的條件內重復讀取非易失性字段可能會導致無限循環，因為編譯器優化可能會將此字段訪問移出循環。
來源：help.semmle.com
如果你替換static boolean found = false;為volatile static boolean found = false;，第一個循環將起作用，但我不推薦它，因為它會浪費你的CPU時間。您應該考慮使用wait和notify。
下面static boolean found，添加static final Object lock = new Object();兩個while循環并將其替換為
try {
synchronized (lock) {
// we will wait here until we get notified
lock.wait();
}
} catch (InterruptedException e) {
e.printStackTrace();
}
main.found = true添加后也是如此
synchronized (main.lock) {
main.lock.notify();
}
最后，你的代碼應該是這樣的
public class main {
static boolean found;
static final Object lock = new Object();
public static void main(String[] args) {
// TODO Auto-generated method stub
int threadCount = 4; //amount of threads to use
Random rand = new Random();
Searcher[] s_arr = new Searcher[threadCount]; //array of threads
int[] arr = new int[10000]; //array to search through
for (int i = 0; i < arr.length; i++) //randomizing #'s in array
arr[i] = (int) (rand.nextFloat() * 1000);
int randIndex = rand.nextInt(arr.length); //choose random index
arr[randIndex] = -1; //set random index to = -1
for (int i = 0; i < threadCount; i++) { //
s_arr[i] = new Searcher(Arrays.copyOfRange(arr, i * (arr.length/threadCount), (i+1) * (arr.length/threadCount)),
(int) (i), i); //assign subarray for this thread to search through
System.out.println(s_arr[i].wait);
s_arr[i].start();
}
try {
synchronized (lock) {
lock.wait();
}
} catch (InterruptedException e) {
e.printStackTrace();
}
System.out.println("found!");
for (int i = 0; i < threadCount; i++) {
try {
s_arr[i].join(); //wait for the threads in order before continuing
System.out.println("Thread ["+i+"] completed");
} catch (InterruptedException e) {
// TODO Auto-generated catch block
e.printStackTrace();
}
}
System.out.println("All threads stopped, program complete.");
}
}
class Searcher extends Thread {
int[] arr;
int wait;
int index;
public Searcher(int[] arr, int wait, int i) {
this.arr = arr;
this.wait = wait;
this.index = i;
}
@Override
public void run() {
for (int i = 0; i < arr.length; i++) {
if (arr[i] == -1) {
System.out.println("["+index+"] -1 Found at index: "+i);
main.found = true;
synchronized (main.lock) {
main.lock.notify();
}
break;
}
if (main.found) break;
//purposely slow down this thread
try {
Thread.sleep(wait);
} catch (InterruptedException e) {
// TODO Auto-generated catch block
e.printStackTrace();
}
}
System.out.println("["+index+"] has stopped");
}
}
添加回答
舉報
0/150
提交
取消