I'm trying example 1 : git checkout 65893ec as the wiki stated.
If I tune up the concurrency config_info.num_concurr_msgs = 2; (main.c:38) , the program crashed with local protection error.

The error is caused by the wrong implementation of the ring buffer.
i.e. (client.c:53)

        buf_offset = (buf_offset + msg_size) % buf_size;
        buf_ptr +=  buf_offset;

should be changed to

        buf_offset = (buf_offset + msg_size) % buf_size;
        buf_ptr = ib_res.ib_buf + buf_offset;

Otherwise, the buffer is never wrap around and causes an overflow.

In example2, i want to use multi-thread transporting, so i modify the value of the num_threads,but when i run the project in the command line, nothing output in the command line, i forced the process to exit, i found in the log,there is some information about IB config, so the qp connect is success,but nothing tranported.

Hi,
Thanks for the great tutorial.
I have an issue when running the first example on both server and client.

================ IB Echo Server ================

************ Configuraion ************

is_server = true
msg_size = 64
num_concurr_msgs = 1
sock_port = 8080

************ End of Configuraion ************

[ERROR] (setup_ib.c:184:setup_ib: errno: Cannot allocate memory) Failed to create qp
[ERROR] (main.c:39:main: errno: None) Failed to setup IB

================ Run Finished ================

the output is as the following:

gcc -Wall -Werror -O2 -c -o config.o config.c
config.c: In function ‘get_rank’:
config.c:105:5: error: ‘strncpy’ output may be truncated copying 64 bytes from a string of length 64 [-Werror=stringop-truncation]
strncpy (hostname, utsname_buf.nodename, sizeof(hostname));
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
cc1: all warnings being treated as errors

gcc does this for security reason, and it seems that many project have got this warning.[1]
I think we can use strlcpy[2] instead:)
[1]openwall/john#3127 (comment)
[2]https://linux.die.net/man/3/strlcpy
https://stackoverflow.com/questions/2114896/why-are-strlcpy-and-strlcat-considered-insecure

两块ConnectX-3的板子直连，用ib_write_lat测试时延，结果出来的数据没看懂什么意思。
比如1024bytes的t_avg竟然有8299usec，就是8ms，是不是太慢了，而如果去掉-C选项，又变成了2usec，又太快了，这是怎么一回事？从源码里来看，为什么不加-C要除CPU的主频，加了又不除了。
那么从哪里能得到正常的数据呢？还是我看的方法不对？

[root@localhost perftest-master]# ib_write_lat -a -n 5 -F -C

• Waiting for client to connect... *

Requested SQ size might be too big. Try reducing TX depth and/or inline size.
Current TX depth is 1 and inline size is 220 .

                RDMA_Write Latency Test

Dual-port : OFF Device : mlx4_0
Number of qps : 1 Transport type : IB
Connection type : RC Using SRQ : OFF
Mtu : 1024[B]
Link type : Ethernet
GID index : 1
Max inline data : 220[B]
rdma_cm QPs : OFF
Data ex. method : Ethernet

local address: LID 0000 QPN 0x0248 PSN 0xe6be8a RKey 0xe0010101 VAddr 0x007f360b400000
GID: 00:00:00:00:00:00:00:00:00:00:255:255:12:12:12:01
remote address: LID 0000 QPN 0x0253 PSN 0x158390 RKey 0x60010102 VAddr 0x007fe0a4e10000
GID: 00:00:00:00:00:00:00:00:00:00:255:255:12:12:12:02

#bytes #iterations t_min[usec] t_max[usec] t_typical[usec] t_avg[usec] t_stdev[usec] 99% percentile[usec] 99.9% percentile[usec]
2 5 2808.00 4636.00 2809.00 2809.00 1.00 4636.00 4636.00
4 5 2775.50 2780.50 2776.50 2776.50 1.00 2780.50 2780.50
8 5 2674.50 2848.50 2722.50 2722.50 48.00 2848.50 2848.50
16 5 2795.50 2841.50 2810.00 2810.25 14.75 2841.50 2841.50
32 5 2856.50 3028.00 2912.00 2912.25 55.75 3028.00 3028.00
64 5 3091.50 3191.50 3141.00 3141.25 49.75 3191.50 3191.50
128 5 3534.50 3601.50 3557.00 3557.00 22.50 3601.50 3601.50
256 5 5689.50 8023.00 5913.50 5913.50 224.00 8023.00 8023.00
512 5 5727.00 6357.00 5826.50 5826.75 99.75 6357.00 6357.00
1024 5 7834.50 8299.00 8040.00 8040.25 205.75 8299.00 8299.00
2048 5 10110.50 10364.50 10195.00 10195.25 84.75 10364.50 10364.50
4096 5 15141.00 15210.50 15152.50 15152.75 11.75 15210.50 15210.50
8192 5 24677.50 24892.50 24784.50 24784.75 107.25 24892.50 24892.50
16384 5 44256.00 44398.00 44310.50 44310.75 54.75 44398.00 44398.00
32768 5 83104.50 83235.50 83160.00 83160.00 55.50 83235.50 83235.50
65536 5 160902.00 161911.00 160909.00 160909.00 7.00 161911.00 161911.00
131072 5 316419.00 316811.50 316430.50 316430.75 11.75 316811.50 316811.50
262144 5 627500.00 627571.50 627504.50 627504.75 4.75 627571.50 627571.50
524288 5 1249507.00 1249553.00 1249521.00 1249521.25 14.25 1249553.00 1249553.00
1048576 5 2493749.50 2493809.00 2493766.00 2493766.25 16.75 2493809.00 2493809.00
2097152 5 4981971.50 4982092.00 4981986.50 4981986.50 15.00 4982092.00 4982092.00
4194304 5 9958520.50 9959012.50 9958554.50 9958554.50 34.00 9959012.50 9959012.50
8388608 5 19911566.50 19911583.50 19911573.00 19911573.25 6.75 19911583.50 19911583.50

这是加了-C选项，出来的时延是不是太大了。而下面不加的又太小了。。

[root@localhost perftest-master]# ib_write_lat -a -n 5 -F

• Waiting for client to connect... *

Requested SQ size might be too big. Try reducing TX depth and/or inline size.
Current TX depth is 1 and inline size is 220 .

                RDMA_Write Latency Test

Dual-port : OFF Device : mlx4_0
Number of qps : 1 Transport type : IB
Connection type : RC Using SRQ : OFF
Mtu : 1024[B]
Link type : Ethernet
GID index : 1
Max inline data : 220[B]
rdma_cm QPs : OFF
Data ex. method : Ethernet

local address: LID 0000 QPN 0x0246 PSN 0xe0bd4b RKey 0xd0010101 VAddr 0x007f7c0cc00000
GID: 00:00:00:00:00:00:00:00:00:00:255:255:12:12:12:01
remote address: LID 0000 QPN 0x0251 PSN 0xf35b27 RKey 0x50010102 VAddr 0x007f27bc793000
GID: 00:00:00:00:00:00:00:00:00:00:255:255:12:12:12:02

#bytes #iterations t_min[usec] t_max[usec] t_typical[usec] t_avg[usec] t_stdev[usec] 99% percentile[usec] 99.9% percentile[usec]
2 5 1.03 1.49 1.09 1.09 0.06 1.49 1.49
4 5 1.03 1.80 1.04 1.04 0.01 1.80 1.80
8 5 0.99 1.05 1.00 1.00 0.01 1.05 1.05
16 5 1.02 1.24 1.03 1.03 0.00 1.24 1.24
32 5 1.07 1.16 1.09 1.09 0.02 1.16 1.16
64 5 1.14 1.18 1.15 1.15 0.01 1.18 1.18
128 5 1.57 1.96 1.62 1.62 0.05 1.96 1.96
256 5 1.75 1.81 1.77 1.77 0.02 1.81 1.81
512 5 2.11 2.15 2.11 2.11 0.00 2.15 2.15
1024 5 2.82 2.84 2.83 2.83 0.01 2.84 2.84
2048 5 3.72 3.78 3.74 3.74 0.01 3.78 3.78
4096 5 5.54 5.59 5.56 5.56 0.01 5.59 5.59
8192 5 9.13 9.35 9.14 9.14 0.01 9.35 9.35
16384 5 16.27 16.34 16.30 16.30 0.03 16.34 16.34
32768 5 30.67 30.73 30.68 30.68 0.00 30.73 30.73
65536 5 59.41 60.78 59.84 59.84 0.43 60.78 60.78
131072 5 116.64 116.70 116.66 116.66 0.02 116.70 116.70
262144 5 231.34 231.42 231.37 231.37 0.02 231.42 231.42
524288 5 460.72 460.87 460.73 460.73 0.01 460.87 460.87
1048576 5 919.49 919.67 919.49 919.49 0.00 919.67 919.67
2097152 5 1836.98 1837.59 1836.99 1836.99 0.01 1837.59 1837.59
4194304 5 3671.97 3672.00 3671.98 3671.98 0.01 3672.00 3672.00
8388608 5 7341.96 7342.20 7342.08 7342.08 0.11 7342.20 7342.20

Error
d7dd63e#r33301093

f73736e#r33301075

Warning -Werror
f73736e#r33301112

f73736e#r33301115

f73736e#r33301135

Hi, sir, after I enter the command ./rdma-tutorial lsy 1234, I'll get a segmentation fault error output, any help?