Re: Weird tcp performance differences with 2.0 and 2.2 kernels

Andrea Arcangeli (andrea@e-mind.com)
Fri, 12 Feb 1999 17:48:04 +0100 (CET)


On Fri, 12 Feb 1999, David Miller wrote:

>There is no way I will allow this change, because it only trades one
>bug for another. If you do quickack mode instead of HZ/50 you will
>get performance problems when talking to SunOS and many other bsd4.3
>derived stacks.

Ok but the other choice I can agree with is to remove the two line
heuristic completly. I don't like a HZ/50 that can be shorter than the
time between two packet from the other end because in such case we will
wait but we won't avoid the ack. Maybe it works fine for most of you and
it's not a big issue for me (since HZ/50 is a pretty slow time anyway) but
I still don't like it.

Andrea Arcangeli

-
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/