Here is a new AVL patch. It uses normal linked lists in the normal
case, and switches to AVL when the number of VMAs exceeds a certain
threshold: 32 VMAs. It thus combines the good timings of plain
2.1.119 with the guaranteed worst-case O(log n) scalability of the
AVL trees.
Such multiple algorithm dynamic schemes have already been shunned by
most of the people on this list early on in the thread. It's what
Solaris does, and it's not only complex, it's also stupid. Find one
algorithm that works nicely in all cases, so we don't have to maintain
a piece of code which uses two different schemes.
Later,
David S. Miller
davem@dm.cobaltmicro.com
-
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/faq.html