summaryrefslogtreecommitdiff
path: root/libraries/base/System/Timeout.hs
blob: d4ef0932b5b8f5ab326559b7a64ff45b649136c6 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
{-# LANGUAGE Safe #-}
{-# LANGUAGE CPP #-}
{-# LANGUAGE StandaloneDeriving #-}

-------------------------------------------------------------------------------
-- |
-- Module      :  System.Timeout
-- Copyright   :  (c) The University of Glasgow 2007
-- License     :  BSD-style (see the file libraries/base/LICENSE)
--
-- Maintainer  :  libraries@haskell.org
-- Stability   :  experimental
-- Portability :  non-portable
--
-- Attach a timeout event to arbitrary 'IO' computations.
--
-------------------------------------------------------------------------------

module System.Timeout ( timeout ) where

#ifndef mingw32_HOST_OS
import Control.Monad
import GHC.Event           (getSystemTimerManager,
                            registerTimeout, unregisterTimeout)
#endif

import Control.Concurrent
import Control.Exception   (Exception(..), handleJust, bracket,
                            uninterruptibleMask_,
                            asyncExceptionToException,
                            asyncExceptionFromException)
import Data.Unique         (Unique, newUnique)

-- An internal type that is thrown as a dynamic exception to
-- interrupt the running IO computation when the timeout has
-- expired.

newtype Timeout = Timeout Unique deriving (Eq)

-- | @since 3.0
instance Show Timeout where
    show _ = "<<timeout>>"

-- Timeout is a child of SomeAsyncException
-- | @since 4.7.0.0
instance Exception Timeout where
  toException = asyncExceptionToException
  fromException = asyncExceptionFromException

-- |Wrap an 'IO' computation to time out and return @Nothing@ in case no result
-- is available within @n@ microseconds (@1\/10^6@ seconds). In case a result
-- is available before the timeout expires, @Just a@ is returned. A negative
-- timeout interval means \"wait indefinitely\". When specifying long timeouts,
-- be careful not to exceed @maxBound :: Int@.
--
-- The design of this combinator was guided by the objective that @timeout n f@
-- should behave exactly the same as @f@ as long as @f@ doesn't time out. This
-- means that @f@ has the same 'myThreadId' it would have without the timeout
-- wrapper. Any exceptions @f@ might throw cancel the timeout and propagate
-- further up. It also possible for @f@ to receive exceptions thrown to it by
-- another thread.
--
-- A tricky implementation detail is the question of how to abort an @IO@
-- computation. This combinator relies on asynchronous exceptions internally.
-- The technique works very well for computations executing inside of the
-- Haskell runtime system, but it doesn't work at all for non-Haskell code.
-- Foreign function calls, for example, cannot be timed out with this
-- combinator simply because an arbitrary C function cannot receive
-- asynchronous exceptions. When @timeout@ is used to wrap an FFI call that
-- blocks, no timeout event can be delivered until the FFI call returns, which
-- pretty much negates the purpose of the combinator. In practice, however,
-- this limitation is less severe than it may sound. Standard I\/O functions
-- like 'System.IO.hGetBuf', 'System.IO.hPutBuf', Network.Socket.accept, or
-- 'System.IO.hWaitForInput' appear to be blocking, but they really don't
-- because the runtime system uses scheduling mechanisms like @select(2)@ to
-- perform asynchronous I\/O, so it is possible to interrupt standard socket
-- I\/O or file I\/O using this combinator.

timeout :: Int -> IO a -> IO (Maybe a)
timeout n f
    | n <  0    = fmap Just f
    | n == 0    = return Nothing
#ifndef mingw32_HOST_OS
    | rtsSupportsBoundThreads = do
        -- In the threaded RTS, we use the Timer Manager to delay the
        -- (fairly expensive) 'forkIO' call until the timeout has expired.
        --
        -- An additional thread is required for the actual delivery of
        -- the Timeout exception because killThread (or another throwTo)
        -- is the only way to reliably interrupt a throwTo in flight.
        pid <- myThreadId
        ex  <- fmap Timeout newUnique
        tm  <- getSystemTimerManager
        -- 'lock' synchronizes the timeout handler and the main thread:
        --  * the main thread can disable the handler by writing to 'lock';
        --  * the handler communicates the spawned thread's id through 'lock'.
        -- These two cases are mutually exclusive.
        lock <- newEmptyMVar
        let handleTimeout = do
                v <- isEmptyMVar lock
                when v $ void $ forkIOWithUnmask $ \unmask -> unmask $ do
                    v2 <- tryPutMVar lock =<< myThreadId
                    when v2 $ throwTo pid ex
            cleanupTimeout key = uninterruptibleMask_ $ do
                v <- tryPutMVar lock undefined
                if v then unregisterTimeout tm key
                     else takeMVar lock >>= killThread
        handleJust (\e -> if e == ex then Just () else Nothing)
                   (\_ -> return Nothing)
                   (bracket (registerTimeout tm n handleTimeout)
                            cleanupTimeout
                            (\_ -> fmap Just f))
#endif
    | otherwise = do
        pid <- myThreadId
        ex  <- fmap Timeout newUnique
        handleJust (\e -> if e == ex then Just () else Nothing)
                   (\_ -> return Nothing)
                   (bracket (forkIOWithUnmask $ \unmask ->
                                 unmask $ threadDelay n >> throwTo pid ex)
                            (uninterruptibleMask_ . killThread)
                            (\_ -> fmap Just f))
        -- #7719 explains why we need uninterruptibleMask_ above.