> TCP's slow-start and congestion avoidance adapt to this, but dropped
> packets, and feedback induced from such events, are what puts the
> gears in motion.
Thanks, Dave. When you combine this startup adaptation with something
analogous happening at the trellis-coding level, it can be nasty. :-}
It's been awhile since I've analyzed tcpdump output, but the packet loss I
saw appeared at various points in the transmission, not just near the
beginning; perhaps some ongoing adaptation is looking for decreased
congestion? If so, it might be handy to be able to tune it, a suitable
default through /proc maybe.
> trust me when I say that we are better off without that code existing any
> more :-)
Ah ha, so that's why 2.0 worked so well! Back to 2.0.37p11. :-) OK, maybe
not...
Kris
-
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/