Re: TCP: user-setable irtt improperly handled (patch included)

Alan Cox (alan@lxorguk.ukuu.org.uk)
Fri, 31 Jan 1997 19:42:02 +0000 (GMT)


> The current TCP implementation allows a per-route preset of the initial
> retransmit timeout (or is it "initial round trip time"?) for all

Yep

> The main thing why I'd like to have someone have a look at it is that I
> had to add something to "struct sock" in order to make it look clean and
> nice, and thus it will affect networking in general (well, just a bit).

Actually I don't think the extra struct sock field is needed. The patch
looks right in its own way but Im sure it ought to use the current route
for the socket to get the irtt - so that if the socket switches route
eg from ethernet to ax.25 the initial rtt and reset estimate is used.

The 2.1.x tcp has also changed a lot so Im not sure if the patch is
relevant to it. I'll file it to check one day

Alan