>> The _only_ way to increase performance is to decrease the tp->rto. I tried
>> force the kernel to retransmit the packet very fast (sure in less time
>> than the rtt hehehe ;-) and my connection to the University now is
>> perfectly responsive (cool ;-). I don' t stall anymore. Before my hack to
>> tcp_reset_xmit_timer() I had to wait 10/15 sec before ssh asks me the
>> password. Now I get the password prompt after some sec (and my RX modem
>> line is always ON ;-). The interactive performance are perfect now with my
>> brute hack.
>
>You've forgotten one point.
>If all people hacked their kernels as you did you aren't able to connect
>to your University at all.
I was aware of that. It was only an exciting hack ;-). Something that I
really would not be able to do on other proprietary OSes ;-). (note also
that the machine crashed after some minutes (just to discurage anyone out
there that would try to reproduce my bad funny idea ;-)).
>Your changes increase the performance of your connection by the price
>of much more congestion for other connections.
This is the reason I have not posted the interesting diff here.
>The retransmit rules were designed for better total performance of
>all connections. The retransmit timeout is set to a value
Yes. Agreed. Using the _right_ retransmit timeout we are _all_ faster.
This is sure.
>giving a very low probability of considering a traveling packet as lost.
>As far as I remember the value of the timeout is set to
>round trip time plus its mean deviation multiplied by 4.
I don' t know what is mean deviation. Can you give me the whole formula so
I' ll do the calc by hand and I' ll see the value calculated by the
kernel, just to be sure?
>What you've done is considered as network abuse in many places
>and it's a good way to be disallowed to use the network.
Here nobody is able to detect such abuse, be sure. I won' t use my brute
TCP hack anymore for other reasons (I deleted my hacked tree now, btw).
Now I' ll only verify that the rto got calculated right from the kernel.
Andrea[s] 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/faq.html