There's been an interesting paper on some recent improvements being made in the Glasgow Haskell Compiler making the rounds on the internet, and I thought some parts might be of interest here. I'm also curious if letting the compiler take care of prefetching is something that might already exist in, say, Intel's C compiler even if it doesn't in gcc.

http://research.microsoft.com/en-us/um/people/simonpj/papers/ndp/haskell-beats-C.pdf