>It could be nice though to send an ack when out of order data arrived, even
>when the data is dropped (to kick fast retransmit on the other side and to
>get the window update there). Currently this is not done. I don't think it
>would solve problems though, just optimize a bit(?)
Fast retransmit then increase cwnd by the 3 packets that are supposed to
be received in the receiver, while instead we have dropped such packets. I
think that fast-retrans is not good to deal with userspace shrunk of the
rcvbuf so probably the best approch is to wait for the rto-expiration as
we was used to do.
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/