Re: [Haskell] [Haskell-cafe] ANN: Monad.Reader Issue 19

2011-10-31 Thread Joey Adams 
On Wed, Oct 26, 2011 at 4:24 PM, Bas van Dijk  wrote:
> I have one question regarding your use of atomicModifyIORef:
>
>  x <- atomicModifyIORef ref (\_ -> (tmstr, ()))
>  x `seq` return ()
>
> Can't you write that as just: writeIORef ref tmstr? If you're not
> using the previous value of the IORef there's no chance of
> inconsistency.

>From the documentation at
http://hackage.haskell.org/packages/archive/base/latest/doc/html/Data-IORef.html
:

IORef operations may appear out-of-order to another thread, ...

...

atomicModifyIORef acts as a barrier to reordering. Multiple
atomicModifyIORef operations occur in strict program order.

Based on this description, it seems that atomicModifyIORef is safer
for writing to an IORef than writeIORef when there are multiple
threads reading and writing it.  If my assessment is correct, I think
Data.IORef should have an atomicWriteIORef :: IORef a -> a -> IO ()
function to clarify this.  I'm not completely sure about this myself.
Could someone confirm this?

Moreover, it'd be nice if there were writeIORef' and
atomicModifyIORef' functions that force the value.  Doing so would
help people avoid making the mistake described by the author.  It's a
really common mistake.

- Joey

___
Haskell mailing list
Haskell@haskell.org
http://www.haskell.org/mailman/listinfo/haskell

Re: [Haskell] [Haskell-cafe] ANN: Monad.Reader Issue 19

2011-10-26 Thread Bas van Dijk 
On 26 October 2011 21:17, Brent Yorgey  wrote:
> I am pleased to announce that Issue 19 of The Monad.Reader, a special
> issue on parallelism and concurrency, is now available:

Thanks, I always really enjoy The Monad.Reader.

> Issue 19 consists of the following three articles:
>
>  * Mighttpd – a High Performance Web Server in Haskell
>    by Kazu Yamamoto

Kazu, really interesting article!

I have one question regarding your use of atomicModifyIORef:

  x <- atomicModifyIORef ref (\_ -> (tmstr, ()))
  x `seq` return ()

Can't you write that as just: writeIORef ref tmstr? If you're not
using the previous value of the IORef there's no chance of
inconsistency.

I looked in the git repository of mighttpd2 and it seems that in the
FileCache module we can make a similar change by rewriting:

remover :: IORef Cache -> IO ()
remover ref = do
threadDelay 1000
_ <- atomicModifyIORef ref (\_ -> (M.empty, ()))
remover ref

to:

remover :: IORef Cache -> IO ()
remover ref = forever $ do
threadDelay 1000
writeIORef ref M.empty

Regards,

Bas

___
Haskell mailing list
Haskell@haskell.org
http://www.haskell.org/mailman/listinfo/haskell