On Mon, Mar 02, 2009 at 10:02:58PM +0000, Duncan Coutts wrote:
This problem could probably be fixed by adjusting the definition of getStdRandom like:
getStdRandom :: (StdGen -> (a,StdGen)) -> IO a getStdRandom f = atomicModifyIORef theStdGen $! (swap . f) where swap (v,g) = g `seq` (g,v)
Though even that may not do it. The StdGen data and stdNext functions may not be strict enough.
data StdGen = StdGen Int32 Int32
It's not strict in the Int32 fields. I doubt that it needs to be so lazy.
If I was right a few years ago, making getStdRandom more strict is indeed enough, although it does change the behaviour of x <- randomRIO undefined which probably noone cares about anyway. There is also already a ticket for this and another problem: http://hackage.haskell.org/trac/ghc/ticket/427 Groeten, Remi