Messages in this thread | | | Date | Fri, 4 Sep 1998 04:24:53 -0700 | From | "David S. Miller" <> | Subject | Re: [PATCH] Re: 2.1.xxx makes Electric Fence 22x slower |
| |
From: alan@lxorguk.ukuu.org.uk (Alan Cox) Date: Fri, 4 Sep 1998 13:16:07 +0100 (BST)
> No way, AVL is not going back into the kernel for vmas, ever.
And pray Dave why not ?
Have you benchmarked the results in detail for all three solutions ?
To expand slightly on my other posting (which you should see soon):
1) If it has balancing overhead like AVL it is not going in, this is why Linus and I removed AVL in the first place. Lower latency than AVL in the common case is a must.
2) As shown by the EFence case, we have to make it scale too.
Fuzzy hash is the only method which has been shown to possess both qualities, work towards a solution using that or stay with the current scheme. This is what I know Linus's decision making process is going to work like here.
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
| |