> Umm.. I've been looking at the use of "tcp_reset_xmit_timer()", and it may be
> that we're seeing a timer fight between the delayed acks and retransmissions,
> where both want to use the same timer and they don't like each other messing
> with it..
>
> Anyway, this would indeed have gotten worse in 1.3.73, because that started
> using more of the delayed ack logic (the code isn't new per se, it's just used
> in a new way). That could have made the problem much worse.
>
Unfortunatly, this does not fix the problem. I tried this
patch against a clean 1.3.85. In fact, it seems to make the problem
slightly worse, but I wouldn't swear to that perception.
____________________________________________________________________
Mark E. Levitt
Department of Speech Communication, Syracuse University
E-mail: melevitt@syr.edu
Home Page: http://web.syr.edu/~melevitt
PGP fingerprint = B8 A3 AA A6 0F 83 9A BE F2 7A 19 F9 15 79 FE A4
Public key available from http://web.syr.edu/~melevitt/pgpkey.html
____________________________________________________________________