> On Fri, Jan 22, 1999 at 12:55:29AM -0800, David S. Miller wrote:
> > I like skip lists too, but one fallacy I found in them for kernel
> > usage is that they require a decent and fast random number source.
> > Perhaps you found a suitable solution to this problem in your
> > application?
>
> I don't think the source has to be _that_ good, so a linear feedback
> shift register should do. You can periodically mix in entropy from the
> random device into the LFSR value to keep it "fairly" random. (Say,
> whenever entropy enters the randomness pool),
Can you imagine the VM or the network layer blocking to wait for
the entorpy pool to refill though. The better the random() the better the
spread on the tree, it becomes more important as N grows.
Benjamin Saller Bender <case@appliedtheory.com>
AppliedTheory Communications Software Engineering Group
Any sufficiently advanced technology is indistinguishable from a rigged demo
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/