My name:李冠霖
My NJU ID:221502012
I collaborated with: NAN
I would like to thank/reward these classmates for their help: NAN
This lab took me about 8 hours to do. I did attend the lab session.
Describe Wrap32 and TCPReceiver structure and design.
Wrap32: 这个还是蛮简单的，就是要看懂他这里给的是什么意思，首先第一个 wrap 很简单直接加就行了，记得做类型转换就行。第二个 unwrap 最开始我想的是先用 0xffffffff00000000 去 mask 一下这个 checkpoint 把高32位取出来，然后再去凑最近的是哪一个，但是后面发现corner case太多了，不如直接去算最近的是哪一个。首先把减完的 offset 去减去 (1ULL << 32) / 2 保证了一定是在更近的那一半上面，然后直接去看相差了多少个 1ULL << 32 就行了
TCPReceiver: 这个没有什么说法，就去翻译给的message里面的字节就行了，对着手册写就ok了，但是需要注意的是，这里的 RST 位是需要和 ByteStream 里面的 has_error() 保持一致的，但是由于某些问题没法直接set，下面细说。
Implementation Challenges:
- 首先在继承的时候，他这个bytestream里面的
set_error() 方法在 reassambler 层面不可用，因为首先 reassambler 层面没有提供这个方法的调用，而且 writer() 和 reader() 两个方法返回的还是 const 对象，所以没办法直接调用 set_error()
- 其次，这个算
checkpoint 附近最近的值怎么算我一开始没想清楚。
Remaining Bugs:
Experimental results and performance：
最后运行结果的bash输出如下：
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
| ❯ cmake --build build --target check2
[0/1] cd /home/eric/minnow/build && /usr/local/bin/ctest --output-on-failure --stop-on-failure --timeout 12 -R '^byte_stream_|^reassembler_|^wrapping|^recv' ByteStream throughput: 21.61 Gbit/s
Reassembler throughput: 12.18 Gbit/s
[1/1] cd /home/eric/minnow/build && /usr/local/bin/ctest --output-on-failure --stop-on-failure --timeout 12 -R '^byte_stream_|^reassembler_|^wrapping|^recv'
Test project /home/eric/minnow/build
Start 1: compile with bug-checkers
1/29 Test
Start 3: byte_stream_basics
2/29 Test
Start 4: byte_stream_capacity
3/29 Test
Start 5: byte_stream_one_write
4/29 Test
Start 6: byte_stream_two_writes
5/29 Test
Start 7: byte_stream_many_writes
6/29 Test
Start 8: byte_stream_stress_test
7/29 Test
Start 9: reassembler_single
8/29 Test
Start 10: reassembler_cap
9/29 Test
Start 11: reassembler_seq
10/29 Test
Start 12: reassembler_dup
11/29 Test
Start 13: reassembler_holes
12/29 Test
Start 14: reassembler_overlapping
13/29 Test
Start 15: reassembler_win
14/29 Test
Start 16: wrapping_integers_cmp
15/29 Test
Start 17: wrapping_integers_wrap
16/29 Test
Start 18: wrapping_integers_unwrap
17/29 Test
Start 19: wrapping_integers_roundtrip
18/29 Test
Start 20: wrapping_integers_extra
19/29 Test
Start 21: recv_connect
20/29 Test
Start 22: recv_transmit
21/29 Test
Start 23: recv_window
22/29 Test
Start 24: recv_reorder
23/29 Test
Start 25: recv_reorder_more
24/29 Test
Start 26: recv_close
25/29 Test
Start 27: recv_special
26/29 Test
Start 37: compile with optimization
27/29 Test
Start 38: byte_stream_speed_test
28/29 Test
Start 39: reassembler_speed_test
29/29 Test
100% tests passed, 0 tests failed out of 29
Total Test time (real) = 13.43 sec
|
图片版本如下：
比较神奇的是这里的reassambler竟然跑的比上一次快。
另外本人写了一个基于 gitlab-ci 的自动化测试工具放在 .gitlab-ci.yml 里面了，只要自己注册个runner就可以跑
Optional: I had unexpected difficulty with: [describe]
Optional: I think you could make this lab better by: [describe]
Optional: I was surprised by: [describe]
Optional: I’m not sure about: [describe]
Optional: I made an extra test I think will be helpful in catching bugs: [describe where to find]
