Yes, that is a sender algorithm.
> Cardwell mentioned in his paper that a part of the improvement resulted
> from the use of usleep() for a more finely-grained timing algorithm.
> Cardwell stated to the effect that a previous kernel implementation
> of Reno-Vegas did not achieve its potential because of the rather
> coarse granularirty of the 10ms (1 jiffy???) "clock tick" that was used....
There is (fortunately) no usleep in kernel.
His patch uses a more finegrained timer to tune some sending decisions,
but this is pure sender side too (if you read the patch you see that
the parts that decide if to send an ACK or not are not touched)
> Isn't Van-Jacobsen a header compression algorithm?
> (I haven't read RFC-1122...)
Van Jacobsen invented lots of algorithms, among them a header compression
algorithm and the standard congestion avoidance in use with TCP nowadays
>
> I'd like to try out your new code. Anything to improve this
> dial-up link would be a gift from heaven ;-)
One thing you can do now is to experiment with shorter queues.
After the PPP interface is established (you need recent net-tools):
/sbin/ifconfig <ppp-interface> txqueuelen <len>
Good values for a PPP connection may be between 3 and 20, you have to
experiment.
-Andi
-- This is like TV. I don't like TV.- 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/