If enable it, and F-Stack app use `ff_close` to close the connection first while the IPv6 client and sever ips in the same prefix scope, the F-Stack app will not respond the last_ack while receiced `fin+ack` package.
It should be default while handling large concurrent requests, otherwise
it will reduce the throughput.
But if there are only a little, you can set it to less than 100, even to
0, to reduce the delay of the connection.
Since maxhold's default value is 1, it may not be enough for some
applicatins, so increase it to the value of
net.link.ether.inet.maxtries(ARP resolution attempts before returning
error).
In current implementation, cores and ports are one to one mapped, this
PR allowed user to specify the core list used to handle single port. for
example cores 1,2,3 handle port 0 and cores 4,5,6 handle port 1.
- Add `port_list` config to dpdk section (used to specify the enabled port list)
- Add `lcore_list` config to port section(used to specify the core list mapped to that port)
- Delete `nb_ports` config in dpdk section
- Delete `port_mask` config in dpdk section
RX: VLAN strip, CRC strip, IP/TCP/UDP checksum.
TX: IP/TCP/UDP checksum, TSO.
Note that TSO has a problem: when enable TSO,
F-Stack is much slower(Temporarily not resolved).
So TSO is disabled by default.
Edit config.ini to enable it.