6 Sep
2009
6 Sep
'09
8:49 a.m.
Dan Doel
On Sunday 06 September 2009 2:18:31 am David Menendez wrote:
It turns out, pseq limits the effectiveness of strictness analysis, because it forces the order of evaluation. John Meacham described this pretty well last week in the Haskell' list http://www.haskell.org/pipermail/haskell-prime/2009-August/003006.html.
Interesting. I hadn't thought of this before, but it certainly makes sense.
Thank to all of you! This thread is fascinating! :) I used `seq` to duct tape my space leaks and stack overflow issues. Now looked for `pseq` for parallelism. And I think I understand enough to use both reasonably. Thanks! Gracjan