Haskell 并发 IO 行为

Question

我试图更深入地了解 Haskell 中的并发性。我有以下代码:

import Control.Concurrent
    main :: IO ()

    main = do 
    arr <- return $ [1..9]
    t <- newMVar 1
    forkIO (takeMVar t >> (print.show) arr >> putMVar t 1)
    forkIO (takeMVar t >> (print.show) arr >> putMVar t 1)
    forkIO (takeMVar t >> (print.show) arr >> putMVar t 1)
    forkIO (takeMVar t >> (print.show) arr >> putMVar t 1)
    forkIO (takeMVar t >> (print.show) arr >> putMVar t 1)
    return ()

有时我发现打印操作重叠，并且得到以下结果(查看第二次调用):

*Main Control.Concurrent> :l test.hs 
[1 of 1] Compiling Main             ( test.hs, interpreted )
Ok, modules loaded: Main.
*Main Control.Concurrent> main
"[1,2,3,4,5,6,7,8,9]"
"[1,2,3,4,5,6,7,8,9]"
"[1,2,3,4,5,6,7,8,9]"
"[1,2,3,4,5,6,7,8,9]"
"[1,2,3,4,5,6,7,8,9]"
*Main Control.Concurrent> main
"[1,2,3,4,5,6,7,8,9]"
"[1,2,3,4,5,6,7,8,9]"
"[1,2,3,4,5,6,7,8,9]"
["[1,2,3,4,5,6,7,8,9]"
?"[1,2,3,4,5,6,7,8,9]"
1h*Main Control.Concurrent> 

我不明白为什么会发生这种情况。将 MVar 用于 [1..9] 效果也不好。

Answer 1

Sometimes i see that the print operations overlaps

输出有时会重叠，因为标准输出上存在锁定。每个线程调用“print”，它尝试写出一个字符，并为每个字符获取锁定。由于 Haskell 的抢占式并发，您将看到线程打印的随机交错。

如果这是不良行为，则让每个线程将其希望打印的字符串发送到打印机线程，然后打印机线程将依次执行逐行输出。