一、什么是柵欄函數(shù)

dispatch_barrier_async函數(shù)，在進程管理中起到一個柵欄的作用，它等待所有位于barrier函數(shù)之前的操作執(zhí)行完畢后執(zhí)行,并且在barrier函數(shù)執(zhí)行之后,barrier函數(shù)之后的操作才會得到執(zhí)行，該函數(shù)需要同dispatch_queue_create函數(shù)生成的DISPATCH_QUEUE_CONCURRENT隊列一起使用。

二、dispatch_barrier_async函數(shù)的作用

1. 實現(xiàn)高效率的數(shù)據(jù)庫訪問和文件訪問
2. 避免數(shù)據(jù)競爭

三、dispatch_barrier_async實例

dispatch_queue_t queue = dispatch_queue_create("testqueue", DISPATCH_QUEUE_CONCURRENT);
    
    dispatch_async(queue, ^{
        NSLog(@"----1-----%@", [NSThread currentThread]);
    });
    dispatch_async(queue, ^{
        NSLog(@"----2-----%@", [NSThread currentThread]);
    });
    
    dispatch_barrier_async(queue, ^{
        NSLog(@"----barrier-----%@", [NSThread currentThread]);
    });
    
    dispatch_async(queue, ^{
        NSLog(@"----3-----%@", [NSThread currentThread]);
    });
    dispatch_async(queue, ^{
        NSLog(@"----4-----%@", [NSThread currentThread]);
    });

輸出結(jié)果

----2-----<NSThread: 0x6000016ef880>{number = 3, name = (null)}
----1-----<NSThread: 0x6000016ef8c0>{number = 4, name = (null)}
----barrier-----<NSThread: 0x6000016ef8c0>{number = 4, name = (null)}
----3-----<NSThread: 0x6000016ef8c0>{number = 4, name = (null)}
----4-----<NSThread: 0x6000016ef880>{number = 3, name = (null)}

四、dispatch_barrier_async和dispatch_barrier_sync的區(qū)別

1. dispatch_barrier_sync

dispatch_queue_t queue = dispatch_queue_create("testqueue", DISPATCH_QUEUE_CONCURRENT);
    NSLog(@"start");
    dispatch_async(queue, ^{
        NSLog(@"---test1---");
    });
    dispatch_async(queue, ^{
        NSLog(@"---test12---");
    });
    
    dispatch_barrier_sync(queue, ^{
        NSLog(@"----barrier-----");
    });
    
    NSLog(@"aaa");
    
    dispatch_async(queue, ^{
        NSLog(@"---test3---");
    });
    dispatch_async(queue, ^{
        NSLog(@"---test4---");
    });
    NSLog(@"last");

運行結(jié)果

start
---test1---
---test12---
----barrier-----
aaa
last
---test3---
---test4---

2. dispatch_barrier_sync

dispatch_queue_t queue = dispatch_queue_create("testqueue", DISPATCH_QUEUE_CONCURRENT);
    NSLog(@"start");
    dispatch_async(queue, ^{
        NSLog(@"---test1---");
    });
    dispatch_async(queue, ^{
        NSLog(@"---test12---");
    });
    
    dispatch_barrier_async(queue, ^{
        NSLog(@"----barrier-----");
    });
    
    NSLog(@"aaa");
    
    dispatch_async(queue, ^{
        NSLog(@"---test3---");
    });
    dispatch_async(queue, ^{
        NSLog(@"---test4---");
    });
    NSLog(@"last");

運行結(jié)果

start
aaa
last
---test1---
---test12---
----barrier-----
---test3---
---test4---

跟sync的情況相比，aaa、last的輸出位置完全不同，async的時候aaa的輸出在test1結(jié)束之前，sync的aaa輸出在test1結(jié)束之后。

3. 總結(jié)

dispatch_barrier_sync(queue,void(^block)())會將queue中barrier前面添加的任務(wù)block全部執(zhí)行后,再執(zhí)行barrier任務(wù)的block,再執(zhí)行barrier后面添加的任務(wù)block.

dispatch_barrier_async(queue,void(^block)())會將queue中barrier前面添加的任務(wù)block只添加不執(zhí)行,繼續(xù)添加barrier的block,再添加barrier后面的block,同時不影響主線程(或者操作添加任務(wù)的線程)中代碼的執(zhí)行!