[En-Nut-Discussion] about tcp stack 2.4 an later

[Harald Kipp]

>
>    Yes, NutEventWait would not timeout. When transmit about 250Kbyte/s,
>it would receive about 33 packets (1.5Kbyte per packet) in 200ms. In other
>word it if one of the sockets's transmiting speed is faster than 7.5KByte/s,
>NutEventWait would not timeout !

You forgot the application side. If large packets
are received, the TCP window would become full soon
and the remote stops sending more packets.

But you are right, if we consider 1 byte data
packets, it might block other connections.


> > Regarding your second example: ARP timeouts won't
> > hurt because of the decoupling mentioned above.
> > Receiving and transmitting is completely asynchronous.
> > The TCP packet without ARP entry is buffered,
> > an ARP request is sent out and the transmit returns,
> > ready to handle new incoming packets. When later
> > on the ARP response is received, the buffered
> > TCP packet will be sent.
>    Maybe I have not understand you

Yes, difficult if two people try to discuss a complicated
topic in a language they both don't know really. :-)

Trying to make it simple: ARP requests won't block
anything. Your second example is wrong, because there's
nothing which will block the thread in the TCP state
machine.

Harald