For exp3, use the following command to generate rip executable file:
g rip.cpp -orip -O2 -lpthread -DRIP_MAIN
# RUN:
./ripThe rip output log may seem like this:
get 192.168. 4. 1 iface enp7s0
get 172. 16. 9. 1 iface vmnet1
get 172. 16.181. 1 iface vmnet8
get 192.168. 3. 2 iface enx000ec6aa5f42
-----------------------------------------------------
| Network | Nexthop Addr | Metric
-----------------------------------------------------
C | 192.168. 4. 1 / 24 | 0. 0. 0. 0 | 1
C | 3. 0. 0. 0 / 24 | 0. 0. 0. 0 | 16
C | 172. 16. 9. 1 / 24 | 0. 0. 0. 0 | 1
C | 172. 16.181. 1 / 24 | 0. 0. 0. 0 | 1
C | 192.168. 3. 2 / 24 | 0. 0. 0. 0 | 1
-----------------------------------------------------
send to 192.168. 4. 1 succeed! packet len: 64
send to 172. 16. 9. 1 succeed! packet len: 64
send to 172. 16.181. 1 succeed! packet len: 64
send to 192.168. 3. 2 succeed! packet len: 64
receive len: 64 cmd: 2 ver: 2 from 192.168.3.1
get 192.168. 4. 1 iface enp7s0
get 172. 16. 9. 1 iface vmnet1
get 172. 16.181. 1 iface vmnet8
get 192.168. 3. 2 iface enx000ec6aa5f42
-----------------------------------------------------
| Network | Nexthop Addr | Metric
-----------------------------------------------------
C | 192.168. 4. 1 / 24 | 0. 0. 0. 0 | 1
C | 3. 0. 0. 0 / 24 | 0. 0. 0. 0 | 16
C | 172. 16. 9. 1 / 24 | 0. 0. 0. 0 | 1
C | 172. 16.181. 1 / 24 | 0. 0. 0. 0 | 1
C | 192.168. 3. 2 / 24 | 0. 0. 0. 0 | 1
R | 192.168. 1. 0 / 24 | 192.168. 3. 1 | 2
R | 192.168.198. 0 / 24 | 192.168. 3. 1 | 3
R | 192.168. 10. 0 / 24 | 192.168. 3. 1 | 3
-----------------------------------------------------
--- 192.168. 1. 0/24 192.168. 3. 1
--- 192.168.198. 0/24 192.168. 3. 1
--- 192.168. 10. 0/24 192.168. 3. 1
send to 192.168. 4. 1 succeed! packet len: 124
send to 172. 16. 9. 1 succeed! packet len: 124
send to 172. 16.181. 1 succeed! packet len: 124
send to 192.168. 3. 2 succeed! packet len: 64
--- 192.168. 1. 0/24 192.168. 3. 1
--- 192.168.198. 0/24 192.168. 3. 1
--- 192.168. 10. 0/24 192.168. 3. 1
receive len: 24 cmd: 1 ver: 2 from 192.168.4.2
send to 192.168. 4. 1 succeed! packet len: 144
--- 192.168. 1. 0/24 192.168. 3. 1
--- 192.168.198. 0/24 192.168. 3. 1
--- 192.168. 10. 0/24 192.168. 3. 1
get 192.168. 4. 1 iface enp7s0
get 172. 16. 9. 1 iface vmnet1
get 172. 16.181. 1 iface vmnet8
get 192.168. 3. 2 iface enx000ec6aa5f42
For exp4, just type mv lookuproute_simple.cpp.bak lookuproute.cpp; make -j and run sudo ./main.
由于为了后续性能，把输出的部分给注释掉了。
编译：直接 make -j
就写了个哈希边表，每次从/32查到/1，看看有没有对应的路由表项
可以理解为我开了32个数组，但是存的时候存得紧凑了点
该算法代码位于 lookuproute.cpp 中，不到30行（后面常数优化了一波，循环展开，做到差不多和4位trie一样快了，代码翻了一倍）
事实上是/32到/13存hash，/12到/1直接数组存下来
如果要单独测试性能的话：
mv test.cpp.bak test.cpp
g test.cpp lookuproute.cpp -oa -O2
# insert 1000 and query 10^7
time ./a 10000000本机i7-8750H差不多0.27s，找了个i7-7700HQ大概0.45s