fix help page

help page
changed default mode to mode 0;improve help page
2025-09-17 20:54:33 +08:00 · 2017-11-19 06:11:06 -06:00 · 2017-11-19 06:07:40 -06:00 · 2017-11-19 06:02:20 -06:00 · 2017-11-19 02:02:44 -08:00 · 2017-11-18 05:31:01 -08:00
21 changed files with 2239 additions and 1907 deletions
--- a/README.md
+++ b/README.md
@@ -1,7 +1,8 @@
 # UDPspeeder
 Network Speed-Up Tool. Boost your Connection on a High Lantency High Packet-Loss Link by using Forward Error Correction.

-When used alone,UDPspeeder speeds-up only UDP connection.Nevertheless,if you used UDPspeeder + any UDP-based VPN together,you can speed-up any traffic(include TCP/UDP/ICMP),currently OpenVPN/L2TP/ShadowVPN are confirmed to be supported。
+When used alone, UDPspeeder speeds-up only UDP connection. Nevertheless, if you used UDPspeeder + any UDP-based VPN together,
+you can speed-up any traffic(include TCP/UDP/ICMP), currently OpenVPN/L2TP/ShadowVPN are confirmed to be supported。

 ![](/images/en/udpspeeder.PNG)

@@ -9,11 +10,22 @@ or

 ![image_vpn](/images/en/udpspeeder+openvpn3.PNG)

+Assume your local network to your server is lossy. Just establish a VPN connection to your server with UDPspeeder + any UDP-based VPN, access your server via this VPN connection, then your connection quality will be significantly improved. With well-tuned parameters , you can easily reduce IP or UDP/ICMP packet-loss-rate to less than 0.01% . Besides reducing packet-loss-rate, UDPspeeder can also significantly improve your TCP latency and TCP single-thread download speed.
+
 [简体中文](/doc/README.zh-cn.md)(内容更丰富)

 ###### Note
-You can use udp2raw with UDPspeeder together to bypass UDP firewalls.
-udp2raw：https://github.com/wangyu-/udp2raw-tunnel
+You can use udp2raw with UDPspeeder together to get better speed on some ISP with UDP QoS(UDP throttling).
+
+udp2raw's repo：
+
+https://github.com/wangyu-/udp2raw-tunnel
+
+You can also try tinyFecVPN, a lightweight high-performance VPN with build-in FEC support:
+
+tinyFecVPN's repo:
+
+https://github.com/wangyu-/tinyFecVPN


 # Efficacy
@@ -26,13 +38,13 @@ tested on a link with 100ms latency and 10% packet loss at both direction
 ![](/images/en/scp_compare2.PNG)

 # Supported Platforms
-Linux host (including desktop Linux,Android phone/tablet,OpenWRT router,or Raspberry PI).
+Linux host (including desktop Linux,Android phone/tablet, OpenWRT router, or Raspberry PI).

-For Windows and MacOS You can run UDPspeeder inside [this](https://github.com/wangyu-/udp2raw-tunnel/releases/download/20170918.0/lede-17.01.2-x86_virtual_machine_image_with_udp2raw_pre_installed.zip) 7.5mb virtual machine image.
+For Windows and MacOS You can run UDPspeeder inside [this](https://github.com/wangyu-/udp2raw-tunnel/releases/download/20171108.0/lede-17.01.2-x86_virtual_machine_image.zip) 7.5mb virtual machine image.

 # How does it work

-UDPspeeder uses FEC(Forward Error Correction) to reduce packet loss rate,at the cost of addtional bandwidth.The algorithm for FEC is called Reed-Solomon.
+UDPspeeder uses FEC(Forward Error Correction) to reduce packet loss rate, at the cost of addtional bandwidth. The algorithm for FEC is called Reed-Solomon.

 ![image0](/images/en/fec.PNG)

@@ -66,15 +78,15 @@ Assume your server ip is 44.55.66.77, you have a service listening on udp port 7
 ./speederv2 -c -l0.0.0.0:3333  -r44.55.66.77:4096 -f20:10 -k "passwd"
 ```

-Now connecting to UDP port 3333 at the client side is equivalent to connecting to port 7777 at the server side,and the connection is boosted by UDPspeeder.
+Now connecting to UDP port 3333 at the client side is equivalent to connecting to port 7777 at the server side, and the connection is boosted by UDPspeeder.

 ##### Note

 `-f20:10` means sending 10 redundant packets for every 20 original packets.

-`-k` enables simple XOR encryption to confuse DPI(Deep Packet Inspection)
+`-k` enables simple XOR encryption

-To run stably,pay attention to MTU.
+To run stably, pay attention to MTU. You can also try `--mode 0`,there will be no MTU problem in this mode.

 # Advanced Topic
 ### Full Options
@@ -125,14 +137,14 @@ log and help options:

 ```
 #### `--fifo` option
-Use a fifo(named pipe) for sending commands to the running program. For example `--fifo fifo.file`,you can use following commands to change parameters dynamically:
+Use a fifo(named pipe) for sending commands to the running program. For example `--fifo fifo.file`, you can use following commands to change parameters dynamically:
 ```
-echo fec 19:9 >fifo.file
-echo mtu 1100 >fifo.file
-echo timeout 5 >fifo.file
-echo queue-len 100 >fifo.file
-echo mode 0 >fifo.file
+echo fec 19:9 > fifo.file
+echo mtu 1100 > fifo.file
+echo timeout 5 > fifo.file
+echo queue-len 100 > fifo.file
+echo mode 0 > fifo.file
 ```
-### Speed-Up any traffic with OpenVPN+UDPspeeder
+### Speed-Up any traffic with OpenVPN + UDPspeeder

 Check [UDPspeeder + openvpn config guide](/doc/udpspeeder_openvpn.md).
--- a/common.cpp
+++ b/common.cpp
@@ -15,12 +15,16 @@ int about_to_exit=0;
 raw_mode_t raw_mode=mode_faketcp;
 unordered_map<int, const char*> raw_mode_tostring = {{mode_faketcp, "faketcp"}, {mode_udp, "udp"}, {mode_icmp, "icmp"}};

-int delay_capacity=0;
+
 //static int random_number_fd=-1;
 char iptables_rule[200]="";
 //int is_client = 0, is_server = 0;

-program_mode_t program_mode=unset_mode;//0 unset; 1client 2server
+program_mode_t client_or_server=unset_mode;//0 unset; 1client 2server
+
+working_mode_t working_mode=tunnel_mode;
+
+int socket_buf_size=1024*1024;



@@ -256,6 +260,39 @@ unsigned short csum(const unsigned short *ptr,int nbytes) {

    return(answer);
 }
+
+
+unsigned short tcp_csum(const pseudo_header & ph,const unsigned short *ptr,int nbytes) {//works both for big and little endian
+
+	register long sum;
+    unsigned short oddbyte;
+    register short answer;
+
+    sum=0;
+	unsigned short * tmp= (unsigned short *)&ph;
+	for(int i=0;i<6;i++)
+	{
+		sum+=*tmp++;
+	}
+
+
+    while(nbytes>1) {
+        sum+=*ptr++;
+        nbytes-=2;
+    }
+    if(nbytes==1) {
+        oddbyte=0;
+        *((u_char*)&oddbyte)=*(u_char*)ptr;
+        sum+=oddbyte;
+    }
+
+    sum = (sum>>16)+(sum & 0xffff);
+    sum = sum + (sum>>16);
+    answer=(short)~sum;
+
+    return(answer);
+}
+
 int set_buf_size(int fd,int socket_buf_size,int force_socket_buf)
 {
 	if(force_socket_buf)
@@ -526,3 +563,62 @@ int create_fifo(char * file)
 	return fifo_fd;
 }

+
+int new_listen_socket(int &fd,u32_t ip,int port)
+{
+	fd =socket(AF_INET, SOCK_DGRAM, IPPROTO_UDP);
+
+	int yes = 1;
+	//setsockopt(udp_fd, SOL_SOCKET, SO_REUSEADDR, &yes, sizeof(yes));
+
+	struct sockaddr_in local_me={0};
+
+	socklen_t slen = sizeof(sockaddr_in);
+	//memset(&local_me, 0, sizeof(local_me));
+	local_me.sin_family = AF_INET;
+	local_me.sin_port = htons(port);
+	local_me.sin_addr.s_addr = ip;
+
+	if (bind(fd, (struct sockaddr*) &local_me, slen) == -1) {
+		mylog(log_fatal,"socket bind error\n");
+		//perror("socket bind error");
+		myexit(1);
+	}
+	setnonblocking(fd);
+    set_buf_size(fd,socket_buf_size);
+
+    mylog(log_debug,"local_listen_fd=%d\n,",fd);
+
+	return 0;
+}
+int new_connected_socket(int &fd,u32_t ip,int port)
+{
+	char ip_port[40];
+	sprintf(ip_port,"%s:%d",my_ntoa(ip),port);
+
+	struct sockaddr_in remote_addr_in = { 0 };
+
+	socklen_t slen = sizeof(sockaddr_in);
+	//memset(&remote_addr_in, 0, sizeof(remote_addr_in));
+	remote_addr_in.sin_family = AF_INET;
+	remote_addr_in.sin_port = htons(port);
+	remote_addr_in.sin_addr.s_addr = ip;
+
+	fd = socket(AF_INET, SOCK_DGRAM, IPPROTO_UDP);
+	if (fd < 0) {
+		mylog(log_warn, "[%s]create udp_fd error\n", ip_port);
+		return -1;
+	}
+	setnonblocking(fd);
+	set_buf_size(fd, socket_buf_size);
+
+	mylog(log_debug, "[%s]created new udp_fd %d\n", ip_port, fd);
+	int ret = connect(fd, (struct sockaddr *) &remote_addr_in, slen);
+	if (ret != 0) {
+		mylog(log_warn, "[%s]fd connect fail\n",ip_port);
+		close(fd);
+		return -1;
+	}
+	return 0;
+}
+
--- a/common.h
+++ b/common.h
@@ -44,7 +44,7 @@
 #include <stdarg.h>
 #include <assert.h>
 #include <linux/if_packet.h>
-
+#include <linux/if_tun.h>

 #include<unordered_map>
 #include<unordered_set>
@@ -67,10 +67,11 @@ typedef u64_t my_time_t;
 const int max_data_len=2200;
 const int buf_len=max_data_len+200;

-const u32_t conv_clear_interval=200;
+
 //const u32_t timer_interval=400;
 ////const u32_t conv_timeout=180000;
-const u32_t conv_timeout=40000;//for test
+//const u32_t conv_timeout=40000;//for test
+const u32_t conv_timeout=180000;
 const int max_conv_num=10000;
 const int max_conn_num=200;

@@ -85,12 +86,12 @@ const u32_t client_retry_interval=1000;

 const u32_t server_handshake_timeout=10000;// this should be much longer than clients. client retry initially ,server retry passtively*/

-const int conv_clear_ratio=10;  //conv grabage collecter check 1/10 of all conv one time
-const int conn_clear_ratio=10;
-const int conv_clear_min=5;
+const int conv_clear_ratio=30;  //conv grabage collecter check 1/30 of all conv one time
+const int conn_clear_ratio=50;
+const int conv_clear_min=1;
 const int conn_clear_min=1;

-
+const u32_t conv_clear_interval=1000;
 const u32_t conn_clear_interval=1000;


@@ -98,7 +99,7 @@ const i32_t max_fail_time=0;//disable

 const u32_t heartbeat_interval=1000;

-const u32_t timer_interval=500;//this should be smaller than heartbeat_interval and retry interval;
+const u32_t timer_interval=400;//this should be smaller than heartbeat_interval and retry interval;

 //const uint32_t conv_timeout=120000; //120 second
 //const u32_t conv_timeout=120000; //for test
@@ -115,10 +116,13 @@ extern int about_to_exit;
 enum raw_mode_t{mode_faketcp=0,mode_udp,mode_icmp,mode_end};
 extern raw_mode_t raw_mode;
 enum program_mode_t {unset_mode=0,client_mode,server_mode};
-extern program_mode_t program_mode;
+extern program_mode_t client_or_server;
 extern unordered_map<int, const char*> raw_mode_tostring ;

-extern int delay_capacity;
+enum working_mode_t {unset_working_mode=0,tunnel_mode,tun_dev_mode};
+extern working_mode_t working_mode;
+
+extern int socket_buf_size;


 typedef u32_t id_t;
@@ -131,8 +135,8 @@ typedef u64_t anti_replay_seq_t;

 typedef u64_t fd64_t;

-enum dest_type{none=0,type_ip_port,type_fd64,type_ip_port_conv,type_fd64_conv/*,type_fd*/};
-
+//enum dest_type{none=0,type_fd64_ip_port,type_fd64,type_fd64_ip_port_conv,type_fd64_conv/*,type_fd*/};
+enum dest_type{none=0,type_fd64_ip_port,type_fd64,type_fd,type_write_fd,type_fd_ip_port/*,type_fd*/};

 struct ip_port_t
 {
@@ -143,11 +147,22 @@ struct ip_port_t
 	char * to_s();
 };

+struct fd64_ip_port_t
+{
+	fd64_t fd64;
+	ip_port_t ip_port;
+};
+struct fd_ip_port_t
+{
+	int fd;
+	ip_port_t ip_port;
+};
 union inner_t
 {
-	ip_port_t ip_port;
-	//int fd;
 	fd64_t fd64;
+	int fd;
+	fd64_ip_port_t fd64_ip_port;
+	fd_ip_port_t fd_ip_port;
 };
 struct dest_t
 {
@@ -162,6 +177,13 @@ struct fd_info_t
 	ip_port_t ip_port;
 };

+struct pseudo_header {
+    u_int32_t source_address;
+    u_int32_t dest_address;
+    u_int8_t placeholder;
+    u_int8_t protocol;
+    u_int16_t tcp_length;
+};

 u64_t get_current_time();
 u64_t get_current_time_us();
@@ -195,6 +217,7 @@ void setnonblocking(int sock);
 int set_buf_size(int fd,int socket_buf_size,int force_socket_buf=0);

 unsigned short csum(const unsigned short *ptr,int nbytes);
+unsigned short tcp_csum(const pseudo_header & ph,const unsigned short *ptr,int nbytes);

 void  signal_handler(int sig);
 int numbers_to_char(id_t id1,id_t id2,id_t id3,char * &data,int &len);
@@ -217,4 +240,8 @@ int create_fifo(char * file);
 int create_new_udp(int &new_udp_fd,int remote_address_uint32,int remote_port);
 */

+int new_listen_socket(int &fd,u32_t ip,int port);
+
+int new_connected_socket(int &fd,u32_t ip,int port);
+
 #endif /* COMMON_H_ */
--- a/connection.cpp
+++ b/connection.cpp
@@ -47,7 +47,7 @@ void conv_manager_t::clear()
 {
 	//if(disable_conv_clear) return ;//////what was the purpose of this code?

-	if(program_mode==server_mode)
+	if(client_or_server==server_mode)
 	{
 		for(auto it=conv_to_u64.begin();it!=conv_to_u64.end();it++)
 		{
@@ -111,7 +111,7 @@ int conv_manager_t::erase_conv(u32_t conv)
 	//if(disable_conv_clear) return 0;
 	assert(conv_last_active_time.find(conv)!=conv_last_active_time.end());
 	u64_t u64=conv_to_u64[conv];
-	if(program_mode==server_mode)
+	if(client_or_server==server_mode)
 	{
 		server_clear_function(u64);
 	}
--- a/connection.h
+++ b/connection.h
@@ -107,7 +107,7 @@ struct conn_info_t     //stores info for a raw connection.for client ,there is o
 	fec_encode_manager_t fec_encode_manager;
 	fec_decode_manager_t fec_decode_manager;
 	my_timer_t timer;
-	ip_port_t ip_port;
+	//ip_port_t ip_port;
 	u64_t last_active_time;
 	stat_t stat;
 	conn_info_t()
--- a/delay_manager.cpp
+++ b/delay_manager.cpp
@@ -66,7 +66,11 @@ int delay_manager_t::add(my_time_t delay,const dest_t &dest,char *data,int len)

 	delay_data_t tmp=delay_data;
 	tmp.data=(char *)malloc(delay_data.len+100);
-
+    if(!tmp.data)
+    {
+        mylog(log_warn, "malloc() returned null in delay_manager_t::add()");
+        return -1;
+    }
 	memcpy(tmp.data,data,delay_data.len);

 	my_time_t tmp_time=get_current_time_us();
--- a/doc/README.zh-cn.md
+++ b/doc/README.zh-cn.md
@@ -11,13 +11,25 @@

 ![image0](/images/Capture2.PNG)

-另外，可以和udp2raw串联使用，在加速的同时把UDP伪装成TCP，防止UDP被运营商QOS或屏蔽。udp2raw: https://github.com/wangyu-/udp2raw-tunnel
+对于某些运营商，UDPspeeder跟udp2raw配合可以达到更好的速度，udp2raw负责把UDP伪装成TCP，来绕过运营商的UDP限速。
+
+udp2raw的repo:
+
+https://github.com/wangyu-/udp2raw-tunnel
+
+如果你嫌UDPspeeder+OpenVPN麻烦，你可以尝试tinyFecVPN，一个集成了UDPspeeder功能的VPN：
+
+tinyFecVPN的repo:
+
+https://github.com/wangyu-/tinyFecVPN
+
+
 #### 效果
-![image0](/images/Capture8.PNG)
+![image0](/images/cn/ping_compare_cn.PNG)

 ![image0](/images/cn/scp_compare.PNG)
 #### 原理简介
-主要原理是通过冗余数据来对抗网络的丢包，发送冗余数据的方式支持FEC(Forward Error Correction)和多倍发包,其中FEC算法是Reed-Solomon。
+主要原理是通过冗余数据来对抗网络的丢包，发送冗余数据的方式支持FEC(Forward Error Correction)和多倍发包，其中FEC算法是Reed-Solomon。

 FEC方式的原理图:

@@ -28,7 +40,7 @@ FEC方式的原理图:

 在多个冗余包之间引入延迟（时间可配）来对抗突发性的丢包，避开中间路由器因为瞬时buffer长度过长而连续丢掉所有副本。

-模拟一定的延迟抖动（时间可配）,这样上层应用计算出来的RTT方差会更大，以等待后续冗余包的到达，不至于发生在冗余包到达之前就触发重传的尴尬。
+模拟一定的延迟抖动（时间可配），这样上层应用计算出来的RTT方差会更大，以等待后续冗余包的到达，不至于发生在冗余包到达之前就触发重传的尴尬。

 输出UDP收发情况报告，可以看出丢包率。

@@ -39,11 +51,13 @@ client支持多个udp连接，server也支持多个client
 # 简明操作说明

 ### 环境要求
-Linux主机，可以是桌面版，可以是android手机/平板，可以是openwrt路由器，也可以是树莓派。在windows和mac上配合虚拟机可以稳定使用（speeder跑在Linux里，其他应用照常跑在window里，桥接模式测试可用），可以使用[这个](https://github.com/wangyu-/udp2raw-tunnel/releases/download/20170918.0/lede-17.01.2-x86_virtual_machine_image_with_udp2raw_pre_installed.zip)虚拟机镜像，大小只有7.5mb，免去在虚拟机里装系统的麻烦。
+Linux主机，可以是桌面版，可以是android手机/平板，可以是openwrt路由器，也可以是树莓派。Release中提供了`amd64`、`x86`、`arm`、`mips_be`、`mips_le`的预编译binary.
+
+对于windows和mac用户，在虚拟机中可以稳定使用（speeder跑在Linux里，其他应用照常跑在window里，桥接模式测试可用）。可以使用[这个](https://github.com/wangyu-/udp2raw-tunnel/releases/download/20171108.0/lede-17.01.2-x86_virtual_machine_image.zip)虚拟机镜像，大小只有7.5mb，免去在虚拟机里装系统的麻烦；虚拟机自带ssh server，可以scp拷贝文件，可以ssh进去，可以复制粘贴，root密码123456。

 android版需要通过terminal运行。

-###### 注意
+##### 注意
 在使用虚拟机时，建议手动指定桥接到哪个网卡，不要设置成自动。否则可能会桥接到错误的网卡。

 ### 安装
@@ -55,25 +69,25 @@ https://github.com/wangyu-/UDPspeeder/releases
 假设你有一个server，ip为44.55.66.77，有一个服务监听在udp 7777端口。 假设你需要加速本地到44.55.66.77:7777的流量。
 ```
 在server端运行:
-./speederv2 -s -l0.0.0.0:4096 -r127.0.0.1:7777  -f20:10 -k "passwd"
+./speederv2 -s -l0.0.0.0:4096 -r127.0.0.1:7777  -f20:10 -k "passwd" --mode 0

 在client端运行:
-./speederv2 -c -l0.0.0.0:3333 -r44.55.66.77:4096 -f20:10 -k "passwd"
+./speederv2 -c -l0.0.0.0:3333 -r44.55.66.77:4096 -f20:10 -k "passwd" --mode 0
 ```

 现在client和server之间建立起了tunnel。想要连接44.55.66.77:7777，只需要连接 127.0.0.1:3333。来回的所有的udp流量会被加速。

-###### 备注:
+##### 备注:

-f20:10 表示对每20个原始数据发送10个冗余包。
+`-f20:10`表示对每20个原始数据发送10个冗余包。`-f20:10` 和`-f 20:10`都是可以的，空格可以省略，对于所有的单字节option都是如此。对于双字节option，例如后面会提到的`--mode 0`，空格不可以省略。

-k 指定一个字符串，server/client间所有收发的包都会被异或，改变协议特征，防止UDPspeeder的协议被运营商针对。
+`-k` 指定一个字符串，开启简单的异或加密

-###### 注意
+推荐使用`--mode 0`选项，否则你可能需要考虑MTU问题。

-要为UDPspeeder加速的应用设置好MTU(不是在UDPspeeder中，是在被加速的应用中)，建议设置为1200。 
+##### 注意

-另外，如果被加速的应用不能调整MTU，也可以在UDPspeeder中通过`--mode 0 --mtu 1200`设置MTU。关于`--mode 0`和`--mtu`的具体含义请看下文。
+这里推荐的参数是给日常/非游戏情况下使用的；玩游戏请用 [使用经验](https://github.com/wangyu-/UDPspeeder/wiki/使用经验) 里推荐的参数。

 # 进阶操作说明

@@ -128,138 +142,103 @@ log and help options:
 ##### `-f` 选项
 设置fec参数，影响数据的冗余度。
 ##### `--timeout` 选项
-指定fec编码器在编码时候最多可以引入多大的延迟。越高fec越有效率，加速游戏时调低可以降低延迟。
-
+指定fec编码器在编码时候最多可以引入多大的延迟。越高fec越有效率，调低可以降低延迟，但是会牺牲效率。
 #####  `--mode` 选项 和 `--mtu`选项
-fec编码器的工作模式。对于mode 0，编码器会积攒一定数量的packet，然后把他们合并再切成等长的片段（切分长度由--mtu指定）。对于mode 1，编码器不会做任何切分,而是会把packet按最大长度对齐，fec冗余包的长度为对齐后的长度（最大长度）。

-mode 0更省流量，在丢包率正常的情况下效果和mode 1是一样的；mode 1延迟更低，在极高丢包的情况下表现更好。
+简单来说`--mode 0`更省流量，没有mtu问题；`--mode 1`可以稍微降低一点延迟，需要考虑mtu；另外还有个`--mode 0 -q1`模式，多倍发包专用，没有延迟，也没有mtu问题，适合游戏，但是最耗流量。

-mode 0使用起来可以不用关注mtu，因为fec编码器会帮你把包切分到合理的大小；用mode 1时必须合理设置上层应用的mtu。mode 0模式中--mtu选项决定切分的片段的长度，mode 1模式中--mtu选项只起检查作用，如果超过了--mtu指定的值，数据包会被丢弃。
+具体见，https://github.com/wangyu-/UDPspeeder/wiki/mode和mtu选项

-mode 0模式的流量消耗基本完全透明。mode 1因为涉及到数据按最大长度对齐，所以流量消耗不是完全可预期。不过就实际使用来看，mode 1消耗的额外流量不多。
-
-mode 0模式数据包一般不会乱序，除非网络本身有严重乱序；mode 1模式被恢复的数据包可能会乱序，不过UDP本来就允许乱序，对绝大多数应用没有影响。mode 0模式反而可以纠正一些乱序情况。
+对于新手，建议不要纠结这些参数的具体含义，就用我在`使用经验`里推荐的设置，不要乱改参数，尤其是不要改`--mtu`。

 ##### `--report`  选项
-数据发送和接受报告。开启后可以根据此数据推测出包速和丢包率等特征。
+数据发送和接受报告。开启后可以根据此数据推测出包速和丢包率等特征。`--report 10`表示每10秒生成一次报告。

 ##### `-i` 选项
-指定一个时间窗口，长度为n毫秒。同一个fec分组的数据在发送时候会被均匀分散到这n毫秒中。可以对抗突发性的丢包。默认值是0，因为这个功能需要用到时钟，在某些虚拟机里时钟不稳定，可能会导致个别包出现非常大的延迟，所以默认关掉了。这个功能很有用，默认参数效果不理想时可以尝试打开。
+指定一个时间窗口，长度为n毫秒。同一个fec分组的数据在发送时候会被均匀分散到这n毫秒中。可以对抗突发性的丢包。默认值是0，因为这个功能需要用到时钟，在某些虚拟机里时钟不稳定，可能会导致个别包出现非常大的延迟，所以默认关掉了。这个功能很有用，默认参数效果不理想时可以尝试打开，比如用`-i 10`。这个选项的跟通信原理上常说的`交错fec` `交织fec`的原理是差不多的。

 ##### `-j` 选项
 为原始数据的发送，增加一个延迟抖动值。这样上层应用计算出来的RTT方差会更大，以等待后续冗余包的到达，不至于发生在冗余包到达之前就触发重传的尴尬。配合-t选项使用。正常情况下跨国网络本身的延迟抖动就很大，可以不用设-j。这个功能也需要时钟，默认关掉了，不过一般情况应该不需要这个功能。

+-j选项不但可以模拟延迟抖动，也可以模拟延迟。
+
 ##### `--random-drop` 选项
-随机丢包。模拟恶劣的网络环境时使用。如果你的网络现在没有多大丢包，但是你想测试一下高丢包环境下各种FEC参数的表现，可以开这个选项。
+随机丢包。模拟高丢包的网络环境时使用。 `--random-drop`和`-j`选项一起用，可以模拟高延迟（或者高延迟抖动）高丢包的网络，可用于测试FEC参数在各种网络环境下的表现。
+
+##### `-q` 选项
+仅对mode 0模式有用。设置fec编码器的最大队列长度。 比如`-q5`的意思是，在编码器积攒了5个数据包后，就立即发送。合理使用可以改善延迟。在下文的`使用经验`里有提到用`--mode 0 -q1` 来多倍发包。 
+
+`-q`和 `--timeout`的作用类似。`-q`决定fec编码器积攒了多少个数据包之后，立即发送。`--timeout`决定编码器收到第一个数据包以后，最多延迟多少毫秒后发送。
+
+默认值是200，也就是尽可能多得积攒数据。 
+
+建议不要自己调整这个参数，除非是用我在`使用经验`里推荐给你的形式。
+
+#### `--fifo` option
+用fifo(命名管道)向运行中的程序发送command。例如`--fifo fifo.file`，可用的command有：
+```
+echo fec 19:9 > fifo.file
+echo mtu 1100 > fifo.file
+echo timeout 5 > fifo.file
+echo queue-len 100 > fifo.file
+echo mode 0 > fifo.file
+```
+可以动态改变fec编码器参数。可以从程序的log里看到command是否发送成功。
+
+### 以下设置两端必须相同。 

 ##### `-k`选项
 指定一个字符串，server/client间所有收发的包都会被异或，改变协议特征，防止UDPspeeder的协议被运营商针对。

 ##### `--disable-obscure`
-UDPspeeder默认情况下会对每个发出的数据包随机填充和异或一些字节(4~32字节)，这样通过抓包难以发现你发了冗余数据，防止VPS被封。这个功能只是为了小心谨慎，即使你关掉这个功能，基本上也没问题，关掉可以省一些带宽和CPU。
-
-##### `-q,--queue-len`
-编码器在做FEC前最多积攒多少个数据包，只对mode 0有效。除非是使用下文`V2版如何多倍发包`里面提到的用法，不建议改动。
-#### `--fifo` option
-用fifo(命名管道)向运行中的程序发送command. 例如`--fifo fifo.file`,可用的command有：
-```
-echo fec 19:9 >fifo.file
-echo mtu 1100 >fifo.file
-echo timeout 5 >fifo.file
-echo queue-len 100 >fifo.file
-echo mode 0 >fifo.file
-```
-可以动态改变fec编码器参数。
+UDPspeeder默认情况下会对每个发出的数据包随机填充和异或一些字节(4~32字节)，这样通过抓包难以发现你发了冗余数据，防止VPS被封。这个功能只是为了小心谨慎，即使你关掉这个功能，基本上也没问题，关掉可以省一些带宽和CPU。`--disable-obscure`可以关掉这个功能。

 # 使用经验

-### 在FEC和多倍发包之间如何选择
-
-对于游戏，游戏的流量本身不大，延迟很重要，多倍发包是最佳解决方案，多倍发包不会引入额外的延迟。FEC编码器需要先积攒一些数据，才可以做FEC,延迟无法避免；对于多倍发包，没有这个问题，所以没有延迟。
-
-对于其他日常应用（延迟要求一般），在合理配置的情况下，FEC的效果肯定好过多倍发包。不过需要根据网络的最大丢包来配置FEC参数，才能有稳定的效果。如果配置不当，对于--mode 1可能会完全没有效果；对于--mode 0，可能效果会比不用UDPspeeder还差。
-
-对于游戏以外的应用，推荐使用FEC。但是，如果FEC版的默认参数在你那边效果很差，而你又不会调，可以先用多倍发包。
-
-### V2版如何多倍发包
-
-只要在设置-f参数时把x设置为1，fec算法就退化为多倍发包了。例如-f1:1,表示2倍发包，-f1:2表示3倍发包，以此类推。另外可以加上`--mode 0 -q1`参数，防止fec编码器试图积攒和合并数据，获得最低的延迟。
-
-2倍发包的完整参数：
-
-```
-./speederv2 -s -l0.0.0.0:4096 -r127.0.0.1:7777  -f1:1 -k "passwd" --mode 0 -q1
-./speederv2 -c -l0.0.0.0:3333 -r44.55.66.77:4096 -f1:1 -k "passwd" --mode 0 -q1
-```
-
-使用了`--mode 0 -q1`以后，`--timeout`选项不起作用，所以不用调。
-
-如果你只需要多倍发包，可以直接用回V1版，V1版配置更简单，占用内存更小，而且经过了几个月的考验，很稳定。
-
-### 根据网络丢包合理设置FEC参数
-
-默认的FEC参数为-f20:10，对每20个包，额外发送10个冗余包，也就是1.5倍发包。已经可以适应绝大多数的网络情况了，对于10%的网络丢包，可以降低到0.01%以下；对于20%的网络丢包，可以降低到2.5%。
-
-如果你的网络丢包很低，比如在3%以下，可以尝试调低参数。比如-f20:5，也就是1.2倍发包，这个参数已经足够把3%的丢包降低到0.01%以下了。
-
-如果网络丢包超过20%，需要把-f20:10调大。
-
-如果你实在不会配，那么也可以用回V1版。
-
-### 根据CPU处理能力来调整FEC参数
-
-FEC算法很吃CPU,初次使用建议关注UDPspeeder的CPU占用。如果CPU被打满，可以在冗余度不变的情况下把FEC分组大小调小，否则的话效果可能很差。
-
-比如-f20:10和-f10:5，都是1.5倍的冗余度，而-f20:10的FEC分组大小是30个包，-f10:5的FEC分组大小是15个包。-f20:10更费CPU,但是在一般情况下效果更稳定。把分组调小可以节省CPU。
-
-另外，fec分组大小不宜过大，否则不但很耗CPU,还有其他副作用，建议x+y<50。
-
-### 改变FEC参数而不断线
-
-`--fifo`选项可以在运行时无缝改变FEC参数，无需重启程序，也不会断线。
-
-### 为什么使用之后效果反而变差了？
-
-有可能是你用了mode 0参数，而又没调好参数。
-
-如果你没有使用mode 0，而确实效果变差了，那很可能是因为你的运营商对UDP有限制。一般看视频和下载都是TCP流量，而用UDPspeeder中转后流量变成了UDP流量，如果运营商对UDP做了限制，就可能导致效果比不用还差。用udp2raw可以解决，udp2raw: https://github.com/wangyu-/udp2raw-tunnel
-
-
-### UDPspeeder和BBR/锐速配合
-
-UDPspeeder和BBR/锐速可以配合使用，UDPspeeder工作在IP层负责降低丢包率，BBR/锐速工作在TCP层负责优化拥塞和重传。这种情况下，可以调低UDPspeeder的冗余度，能把丢包率降低到5%以内就可以了，剩下的交给BBR/锐速解决，这样预计可以节省一些流量。如果是UDPspeeder跟Linux默认的Cubic一起用，最少也要把丢包率降低到1%以下才能流畅使用TCP。
-
-对下文的`UDPspeeder + openvpn`和`UDPspeeder + openvpn + $***`方法有效。不过有一点区别，具体见下文。
-
-### UDPspeeder和Kcptun配合
-
-UDPspeeder和Kcptun配合,UDPspeeder和Kcptun可以并联也可以串联。
-
-并联的情况下，让kcptun负责加速TCP,UDPspeeder负责加速UDP。见下文的`UDPspeeder + kcptun + $*** 同时加速tcp和udp流量`。
-
-串联的情况。UDPspeeder的FEC跟Kcptun自带的相比：可以对两个方向设置不同的FEC参数、有一个更省流量的mode 0模式、可以动态改变FEC参数；但是UDPspeeder本身不优化拥塞和重传算法。所以UDPspeeder和Kcptun也可以配合使用，结合两者的优点。
-
-串联时可以关掉Kcptun的FEC,让UDPspeeder接管FEC功能。这样UDPspeeder工作在UDP层负责降低丢包率，Kcptun工作在应用层用kcp算法负责优化拥塞和重传，能起到和`UDPspeeder+BBR/锐速`类似的效果。
-
-如果发Issue问Kcptun+UDPspeeder相关的问题，一定要说明是并联还是串联。
+https://github.com/wangyu-/UDPspeeder/wiki/使用经验

 # 应用

-#### UDPspeeder + openvpn加速任何流量
+#### UDPspeeder + OpenVPN加速任何流量，也适用于其他VPN
 ![image0](/images/Capture2.PNG)

-具体配置见，[UDPspeeder + openvpn config guide](/doc/udpspeeder_openvpn.md).
-
-可以和BBR/锐速叠加，不过BBR/锐速部署在VPS上只对从本地到VPS的流量有效，对从本地到第三方服务器的流量无效。
+可以和BBR/锐速叠加，不过BBR/锐速部署在VPS上只对本地和VPS间的流量有效，对从本地和第三方服务器间的流量无效。

 需要在服务端开启ipforward和NAT。在客户端改路由表（可以手动修改，也可以由OpenVPN的redirect-gateway选项自动加好）。

+Linux具体配置: [UDPspeeder + openvpn config guide](/doc/udpspeeder_openvpn.md).
+
+Windows具体配置: [win10系统UDPspeeder+OpenVPN的完整设置](https://github.com/wangyu-/UDPspeeder/wiki/win10系统UDPspeeder-OpenVPN的完整设置)
+
+如果UDPspeeder + OpenVPN对你来说显得太麻烦了，你可以尝试一下tinyFecVPN,一个集成了UDPspeeder功能的VPN:
+
+https://github.com/wangyu-/tinyFecVPN/
+
 #### UDPspeeder + kcptun/finalspeed + $*** 同时加速tcp和udp流量
-如果你需要用加速的tcp看视频和下载文件，这样效果比UDPspeeder+vpn方案更好（在没有BBR的情况下）。
+如果你需要用加速的tcp看视频和下载文件，这样效果可能比没有BBR的UDPspeeder+vpn方案更好。另外，如果你需要玩游戏，但是嫌配VPN麻烦，也可以用这种方案。
 ![image0](/images/cn/speeder_kcptun.PNG)

-#### UDPspeeder + openvpn + $*** 混合方案
+具体配置方法简介:
+
+假设$\*\*\*  server监听在在44.55.66.77的443端口(tcp和udp同时)。用kcptun把tcp 443映射到本地的tcp 1234；用UDPspeeder把udp 443的映射到本地的udp 1234。
+然后让$\*\*\* client 去连127.0.0.1:1234就可以了，tcp和udp都被加速了。完整命令：
+```
+run at server side:
+./kcp_server  -l ":4000" -t "127.0.0.1:443" -mode fast2
+./speederv2 -s -l0.0.0.0:4001 -r127.0.0.1:443  -f20:10 -k "passwd"
+
+run at client side:
+./kcp_client  -l ":1234" -r "44.55.66.77:4000" -mode fast2
+./speederv2 -c -l0.0.0.0:1234 -r44.55.66.77:4001 -f20:10 -k "passwd"
+```
+
+这就是全部的命令了。issue里有很多人困惑于怎么把tcp和udp流量"分开"，其实很简单就可以做到。
+
+如果只需要加速UDP，不需要加速TCP，可以把kcptun换成其他的任意端口转发方式，比如ncat/socat/ssh tunnel/iptables/[tinyPortMapper](https://github.com/wangyu-/tinyPortMapper/releases)。
+
+另外，如果没有kcptun只有BBR/锐速的话，也可以把kcptun换成ncat/socat/ssh tunnel/iptables/[tinyPortMapper](https://github.com/wangyu-/tinyPortMapper/releases)。这样，TCP流量由锐速/BBR加速，UDP由UDPspeeder加速。
+
+#### UDPspeeder + openvpn + $*** 混合方案，也适用于其他VPN
 也是我正在用的方案。优点是可以随时在vpn和$\*\*\*方案间快速切换。
 实际部署起来比图中看起来的还要简单。不需要改路由表，不需要写iptables规则和开启NAT，需要做的只是用openvpn分配的ip访问$*** server。

@@ -269,6 +248,20 @@ UDPspeeder和Kcptun配合,UDPspeeder和Kcptun可以并联也可以串联。

 可以和BBR/锐速叠加，BBR/锐速只要部署在VPS上就有效。

+也可以用[tinyFecVPN](https://github.com/wangyu-/tinyFecVPN/) + $\*\*\* ，配置起来更简单。
+
+# 应用实例
+
+#### win10系统UDPspeeder+OpenVPN的完整设置
+
+https://github.com/wangyu-/UDPspeeder/wiki/win10系统UDPspeeder-OpenVPN的完整设置
+
+
+#### 用树莓派做路由器，搭建透明代理，加速游戏主机的网络
+
+https://github.com/wangyu-/UDPspeeder/wiki/用树莓派做路由器，搭建透明代理，加速游戏主机的网络
+
+
 # 编译教程
 暂时先参考udp2raw的这篇教程，几乎一样的过程。

--- a/doc/README.zh-cn.v1.md
+++ b/doc/README.zh-cn.v1.md
@@ -15,7 +15,7 @@ UDP双边加速工具，降低丢包率，配合vpn可以加速任何协议，

 每个冗余数据包都是间隔数毫秒（可配置）以后延迟发出的，可以避开中间路由器因为瞬时buffer长度过长而连续丢掉所有副本。

-模拟一定的延迟抖动,这样上层应用计算出来的RTT方差会更大，以等待后续冗余包的到达，不至于发生在冗余包到达之前就触发重传的尴尬。
+模拟一定的延迟抖动，这样上层应用计算出来的RTT方差会更大，以等待后续冗余包的到达，不至于发生在冗余包到达之前就触发重传的尴尬。

 #### 适用场景
 绝大部分流量不高的情况。程序本身加速udp，但是配合openvpn可以加速任何流量。网络状况不好时，游戏卡得没法玩，或者网页卡得没法打开，使用起来效果最好。对于解决语音通话的断断续续效果也不错。不适合大流量的场景，比如BT下载和在线看视频。 无论从自己使用效果的角度，还是从国际出口带宽占用的角度，都建议不要在大流量环境使用。
@@ -134,7 +134,7 @@ other options:

 跟openvpn via kcptun方式的对比：

-kcptun在udp层有RS code，也是一种冗余传输，通过openvpn把流量转成tcp，再通过kcptun加速是有一定效果的。但是tcp只支持按序到达。按序到达的意思是,如果你发了1 2 3 4 5 6 ,6个包，如果第一个包丢了，那么必须等第一个包重传成功以后 2 3 4 5 6 才能到达；只要有一个包不到，后续数据包就要一直等待。用tcp承载udp流量会破坏udp的实时性。会造成游戏卡顿更严重。
+kcptun在udp层有RS code，也是一种冗余传输，通过openvpn把流量转成tcp，再通过kcptun加速是有一定效果的。但是tcp只支持按序到达。按序到达的意思是，如果你发了1 2 3 4 5 6，6个包，如果第一个包丢了，那么必须等第一个包重传成功以后 2 3 4 5 6 才能到达；只要有一个包不到，后续数据包就要一直等待。用tcp承载udp流量会破坏udp的实时性。会造成游戏卡顿更严重。

 udp协议本身是ip协议加上了端口之后的直接封装，udp继承了ip协议的实时/乱序到达特性，更适合中转vpn。

--- a/doc/udpspeeder_openvpn.md
+++ b/doc/udpspeeder_openvpn.md
@@ -39,7 +39,7 @@ mute 20

 comp-lzo no

-fragment 1200       ##### very important    you can turn it up a bit. but,the lower the safer
+fragment 1200       ##### very important    you can turn it up a bit. but, the lower the safer
 mssfix 1200         ##### very important

 sndbuf 2000000      ##### important
@@ -79,7 +79,7 @@ mute 20

 comp-lzo no

-fragment 1200       ##### very important    you can turn it up a bit. but,the lower the safer
+fragment 1200       ##### very important    you can turn it up a bit. but, the lower the safer
 mssfix 1200         ##### very important

 sndbuf 2000000      ##### important
@@ -88,9 +88,9 @@ txqueuelen 4000     ##### suggested
 ```

 ##### Note:
-If you use the `redirect-gateway` option of OpenVPN,you may need to add a route exception for your remote server ip at client side.Otherwise OpenVPN may hijack UDPspeeder 's traffic.
+If you use the `redirect-gateway` option of OpenVPN, you may need to add a route exception for your remote server ip at client side.Otherwise OpenVPN may hijack UDPspeeder 's traffic.

-For example,depend on your network environment,the command may looks like:
+For example, depend on your network environment, the command may looks like:
 ```
 ip route add 44.55.66.77 via 44.55.66.1
 ```
@@ -100,3 +100,8 @@ or
 ip route add 44.55.66.77 dev XXX
 ```
 (run at client side)
+
+##### Other Info
+You can also use tinyFecVPN,a lightweight VPN with build-in FEC support:
+
+https://github.com/wangyu-/tinyFecVPN
--- a/fec_manager.cpp
+++ b/fec_manager.cpp
@@ -16,7 +16,7 @@ int g_fec_redundant_num=10;
 int g_fec_mtu=1250;
 int g_fec_queue_len=200;
 int g_fec_timeout=8*1000; //8ms
-int g_fec_mode=1;
+int g_fec_mode=0;

 int dynamic_update_fec=1;

@@ -146,11 +146,11 @@ fec_encode_manager_t::fec_encode_manager_t()
 	}
 	timer_fd64=fd_manager.create(timer_fd);

-	re_init(fec_data_num,fec_redundant_num,fec_mtu,fec_queue_len,fec_timeout,fec_mode);
+	reset_fec_parameter(g_fec_data_num,g_fec_redundant_num,g_fec_mtu,g_fec_queue_len,g_fec_timeout,g_fec_mode);
+

-	seq=(u32_t)get_true_random_number(); //TODO temp solution for a bug.
 }
-int fec_encode_manager_t::re_init(int data_num,int redundant_num,int mtu,int queue_len,int timeout,int mode)
+int fec_encode_manager_t::reset_fec_parameter(int data_num,int redundant_num,int mtu,int queue_len,int timeout,int mode)
 {
 	fec_data_num=data_num;
 	fec_redundant_num=redundant_num;
@@ -160,15 +160,8 @@ int fec_encode_manager_t::re_init(int data_num,int redundant_num,int mtu,int que
 	fec_mode=mode;

 	assert(data_num+redundant_num<max_fec_packet_num);
-	counter=0;
-	blob_encode.clear();
-	ready_for_output=0;
-	//seq=0;

-	itimerspec zero_its;
-	memset(&zero_its, 0, sizeof(zero_its));
-
-	timerfd_settime(timer_fd, TFD_TIMER_ABSTIME, &zero_its, 0);
+	clear();

 	return 0;
 }
@@ -192,7 +185,7 @@ int fec_encode_manager_t::append(char *s,int len/*,int &is_first_packet*/)
 	{
 		mylog(log_trace,"counter=%d\n",counter);
 		assert(len<=65535&&len>=0);
-		assert(len<=fec_mtu);
+		//assert(len<=fec_mtu);//relax this limitation
 		char * p=input_buf[counter]+sizeof(u32_t)+4*sizeof(char);//copy directly to final position,avoid unnecessary copy.
 		//remember to change this,if protocol is modified

@@ -225,15 +218,19 @@ int fec_encode_manager_t::input(char *s,int len/*,int &is_first_packet*/)
 	//int counter_back=counter;
 	assert(fec_mode==0||fec_mode==1);

-	if(fec_mode==0&& s!=0 &&counter==0&&blob_encode.get_shard_len(fec_data_num,len)>=fec_mtu)
+	if(fec_mode==0&& s!=0 &&counter==0)
 	{
-		mylog(log_warn,"message too long len=%d,ignored\n",len);
+		int out_len=blob_encode.get_shard_len(fec_data_num,len);
+		if(out_len>fec_mtu)
+		{
+			mylog(log_warn,"message too long ori_len=%d out_len=%d fec_mtu=%d,ignored\n",len,out_len,fec_mtu);
 			return -1;
 		}
-	if(fec_mode==1&&s!=0&&len>=fec_mtu)
+	}
+	if(fec_mode==1&&s!=0&&len>fec_mtu)
 	{
-		mylog(log_warn,"message too long len=%d fec_mtu=%d,ignored\n",len,fec_mtu);
-		return -1;
+		mylog(log_warn,"mode==1,message len=%d,len>fec_mtu,fec_mtu=%d,packet may not be delivered\n",len,fec_mtu);
+		//return -1;
 	}
 	if(s==0&&counter==0)
 	{
@@ -242,7 +239,7 @@ int fec_encode_manager_t::input(char *s,int len/*,int &is_first_packet*/)
 	}
 	if(s==0) about_to_fec=1;//now

-	if(fec_mode==0&& blob_encode.get_shard_len(fec_data_num,len)>=fec_mtu) {about_to_fec=1; delayed_append=1;}//fec then add packet
+	if(fec_mode==0&& blob_encode.get_shard_len(fec_data_num,len)>fec_mtu) {about_to_fec=1; delayed_append=1;}//fec then add packet

 	if(fec_mode==0) assert(counter<fec_queue_len);//counter will never equal fec_pending_num,if that happens fec should already been done.
 	if(fec_mode==1) assert(counter<fec_data_num);
@@ -285,7 +282,10 @@ int fec_encode_manager_t::input(char *s,int len/*,int &is_first_packet*/)
    			int best_data_num=fec_data_num;
    			for(int i=1;i<actual_data_num;i++)
    			{
-    				u32_t new_len=(blob_encode.get_shard_len(i,0)+header_overhead)*(i+fec_redundant_num);
+    				u32_t shard_len=blob_encode.get_shard_len(i,0);
+    				if(shard_len>(u32_t)fec_mtu) continue;
+
+    				u32_t new_len=(shard_len+header_overhead)*(i+fec_redundant_num);
    				if(new_len<best_len)
    				{
    					best_len=new_len;
@@ -489,15 +489,12 @@ int fec_encode_manager_t::output(int &n,char ** &s_arr,int *&len)
 	}
 	return 0;
 }
-
+/*
 int fec_decode_manager_t::re_init()
 {
-	for(int i=0;i<(int)fec_buff_num;i++)
-		fec_data[i].used=0;
-	ready_for_output=0;
-	index=0;
+	clear();
 	return 0;
-}
+}*/

 int fec_decode_manager_t::input(char *s,int len)
 {
--- a/fec_manager.h
+++ b/fec_manager.h
@@ -37,10 +37,16 @@ struct anti_replay_t
 	unordered_set<u32_t> st;
 	int index;
 	anti_replay_t()
+	{
+		clear();
+	}
+	int clear()
 	{
 		memset(replay_buffer,-1,sizeof(replay_buffer));
+		st.clear();
 		st.rehash(anti_replay_buff_size*3);
 		index=0;
+		return 0;
 	}
 	void set_invaild(u32_t seq)
 	{
@@ -140,6 +146,22 @@ public:
 	fec_encode_manager_t();
 	~fec_encode_manager_t();

+	int clear()
+	{
+		counter=0;
+		blob_encode.clear();
+		ready_for_output=0;
+
+		itimerspec zero_its;
+		memset(&zero_its, 0, sizeof(zero_its));
+
+		timerfd_settime(timer_fd, TFD_TIMER_ABSTIME, &zero_its, 0);
+
+		seq=(u32_t)get_true_random_number(); //TODO temp solution for a bug.
+
+		return 0;
+	}
+
 	my_time_t get_first_packet_time()
 	{
 		return first_packet_time_for_output;
@@ -155,7 +177,7 @@ public:
 		return fec_mode;
 	}
 	u64_t get_timer_fd64();
-	int re_init(int data_num,int redundant_num,int mtu,int pending_num,int pending_time,int type);
+	int reset_fec_parameter(int data_num,int redundant_num,int mtu,int pending_num,int pending_time,int type);
 	int input(char *s,int len/*,int &is_first_packet*/);
 	int output(int &n,char ** &s_arr,int *&len);
 };
@@ -200,7 +222,7 @@ public:
 	fec_decode_manager_t()
 	{
 		fec_data=new fec_data_t[fec_buff_num+5];
-		re_init();
+		clear();
 	}
 	fec_decode_manager_t(const fec_decode_manager_t &b)
 	{
@@ -210,7 +232,21 @@ public:
 	{
 		delete fec_data;
 	}
-	int re_init();
+	int clear()
+	{
+		anti_replay.clear();
+		mp.clear();
+		mp.rehash(fec_buff_num*3);
+
+		for(int i=0;i<(int)fec_buff_num;i++)
+			fec_data[i].used=0;
+		ready_for_output=0;
+		index=0;
+
+		return 0;
+	}
+
+	//int re_init();
 	int input(char *s,int len);
 	int output(int &n,char ** &s_arr,int* &len_arr);
 };
--- a/images/cn/ping_compare_cn.PNG
+++ b/images/cn/ping_compare_cn.PNG
--- a/lib/rs.c
+++ b/lib/rs.c
@@ -51,6 +51,10 @@ void* get_code(int k,int n)
 	if (table==0)
 	{
 		table=(void* (*)[256]) malloc(sizeof(void*)*256*256);
+		if(!table)
+		{
+		    return table;
+		}
 		memset(table,0,sizeof(void*)*256*256);
 	}
 	if(table[k][n]==0)
--- a/main.cpp
+++ b/main.cpp
--- a/5
+++ b/5
@@ -8,12 +8,12 @@ cc_arm= /toolchains/arm-2014.05/bin/arm-none-linux-gnueabi-g++
 #cc_bcm2708=/home/wangyu/raspberry/tools/arm-bcm2708/gcc-linaro-arm-linux-gnueabihf-raspbian/bin/arm-linux-gnueabihf-g++ 
 FLAGS= -std=c++11   -Wall -Wextra -Wno-unused-variable -Wno-unused-parameter -Wno-missing-field-initializers -ggdb

-SOURCES=main.cpp log.cpp common.cpp lib/fec.c lib/rs.c packet.cpp delay_manager.cpp fd_manager.cpp connection.cpp fec_manager.cpp
+SOURCES=main.cpp log.cpp common.cpp lib/fec.c lib/rs.c packet.cpp delay_manager.cpp fd_manager.cpp connection.cpp fec_manager.cpp misc.cpp tunnel.cpp

 NAME=speederv2
 TARGETS=amd64 arm mips24kc_be x86  mips24kc_le

-TAR=${NAME}_binaries.tar.gz `echo ${TARGETS}|sed -r 's/([^ ]+)/speederv2_\1/g'`
+TAR=${NAME}_binaries.tar.gz `echo ${TARGETS}|sed -r 's/([^ ]+)/speederv2_\1/g'` version.txt

 all:git_version
 	rm -f ${NAME}
@@ -56,6 +56,7 @@ cross3:git_version
 	${cc_cross}   -o ${NAME}_cross    -I. ${SOURCES} ${FLAGS} -lrt -static -O3

 release: ${TARGETS} 
+	cp git_version.h version.txt
 	tar -zcvf ${TAR}

 clean:	
--- a/misc.cpp
+++ b/misc.cpp
@@ -0,0 +1,967 @@
+/*
+ * misc.cpp
+ *
+ *  Created on: Oct 26, 2017
+ *      Author: root
+ */
+
+
+#include "misc.h"
+
+
+char fifo_file[1000]="";
+
+int mtu_warn=1350;
+
+int disable_mtu_warn=1;
+int disable_fec=0;
+
+int debug_force_flush_fec=0;
+
+
+int jitter_min=0*1000;
+int jitter_max=0*1000;
+
+int output_interval_min=0*1000;
+int output_interval_max=0*1000;
+
+int fix_latency=0;
+
+u32_t local_ip_uint32,remote_ip_uint32=0;
+char local_ip[100], remote_ip[100];
+int local_port = -1, remote_port = -1;
+
+conn_manager_t conn_manager;
+delay_manager_t delay_manager;
+fd_manager_t fd_manager;
+
+int time_mono_test=0;
+
+int delay_capacity=0;
+
+
+char sub_net[100]="10.22.22.0";
+u32_t sub_net_uint32=0;
+
+char tun_dev[100]="";
+
+
+int keep_reconnect=0;
+
+int tun_mtu=1500;
+
+int mssfix=1;
+
+
+int from_normal_to_fec(conn_info_t & conn_info,char *data,int len,int & out_n,char **&out_arr,int *&out_len,my_time_t *&out_delay)
+{
+
+	static my_time_t out_delay_buf[max_fec_packet_num+100]={0};
+	//static int out_len_buf[max_fec_packet_num+100]={0};
+	//static int counter=0;
+	out_delay=out_delay_buf;
+	//out_len=out_len_buf;
+	inner_stat_t &inner_stat=conn_info.stat.normal_to_fec;
+	if(disable_fec)
+	{
+		if(data==0)
+		{
+			out_n=0;
+			return 0;
+		}
+		//assert(data!=0);
+		inner_stat.input_packet_num++;
+		inner_stat.input_packet_size+=len;
+		inner_stat.output_packet_num++;
+		inner_stat.output_packet_size+=len;
+
+		out_n=1;
+		static char *data_static;
+		data_static=data;
+		static int len_static;
+		len_static=len;
+		out_arr=&data_static;
+		out_len=&len_static;
+		out_delay[0]=0;
+
+	}
+	else
+	{
+		if(data!=0)
+		{
+			inner_stat.input_packet_num++;
+			inner_stat.input_packet_size+=len;
+		}
+		//counter++;
+
+		conn_info.fec_encode_manager.input(data,len);
+
+		//if(counter%5==0)
+			//conn_info.fec_encode_manager.input(0,0);
+
+		//int n;
+		//char **s_arr;
+		//int s_len;
+
+
+		conn_info.fec_encode_manager.output(out_n,out_arr,out_len);
+
+		if(out_n>0)
+		{
+			my_time_t common_latency=0;
+			my_time_t first_packet_time=conn_info.fec_encode_manager.get_first_packet_time();
+
+			if(fix_latency==1&&conn_info.fec_encode_manager.get_type()==0)
+			{
+				my_time_t current_time=get_current_time_us();
+				my_time_t tmp;
+				assert(first_packet_time!=0);
+				//mylog(log_info,"current_time=%llu first_packlet_time=%llu   fec_pending_time=%llu\n",current_time,first_packet_time,(my_time_t)fec_pending_time);
+				if((my_time_t)conn_info.fec_encode_manager.get_pending_time() >=(current_time - first_packet_time))
+				{
+					tmp=(my_time_t)conn_info.fec_encode_manager.get_pending_time()-(current_time - first_packet_time);
+					//mylog(log_info,"tmp=%llu\n",tmp);
+				}
+				else
+				{
+					tmp=0;
+					//mylog(log_info,"0\n");
+				}
+				common_latency+=tmp;
+			}
+
+			common_latency+=random_between(jitter_min,jitter_max);
+
+			out_delay_buf[0]=common_latency;
+
+			for(int i=1;i<out_n;i++)
+			{
+				out_delay_buf[i]=out_delay_buf[i-1]+ (my_time_t)( random_between(output_interval_min,output_interval_max)/(out_n-1)  );
+			}
+		}
+
+	}
+
+	mylog(log_trace,"from_normal_to_fec input_len=%d,output_n=%d\n",len,out_n);
+
+	if(out_n>0)
+	{
+		log_bare(log_trace,"seq= %u ",read_u32(out_arr[0]));
+	}
+	for(int i=0;i<out_n;i++)
+	{
+		inner_stat.output_packet_num++;
+		inner_stat.output_packet_size+=out_len[i];
+
+		log_bare(log_trace,"%d ",out_len[i]);
+
+	}
+
+	log_bare(log_trace,"\n");
+	//for(int i=0;i<n;i++)
+	//{
+		//delay_send(0,dest,s_arr[i],s_len);
+	//}
+	//delay_send(0,dest,data,len);
+	//delay_send(1000*1000,dest,data,len);
+	return 0;
+}
+int from_fec_to_normal(conn_info_t & conn_info,char *data,int len,int & out_n,char **&out_arr,int *&out_len,my_time_t *&out_delay)
+{
+	static my_time_t out_delay_buf[max_blob_packet_num+100]={0};
+	out_delay=out_delay_buf;
+	inner_stat_t &inner_stat=conn_info.stat.fec_to_normal;
+	if(disable_fec)
+	{
+		assert(data!=0);
+		inner_stat.input_packet_num++;
+		inner_stat.input_packet_size+=len;
+		inner_stat.output_packet_num++;
+		inner_stat.output_packet_size+=len;
+
+		if(data==0)
+		{
+			out_n=0;
+			return 0;
+		}
+		out_n=1;
+		static char *data_static;
+		data_static=data;
+		static int len_static;
+		len_static=len;
+		out_arr=&data_static;
+		out_len=&len_static;
+		out_delay[0]=0;
+	}
+	else
+	{
+
+		if(data!=0)
+		{
+			inner_stat.input_packet_num++;
+			inner_stat.input_packet_size+=len;
+		}
+
+		conn_info.fec_decode_manager.input(data,len);
+
+		//int n;char ** s_arr;int* len_arr;
+		conn_info.fec_decode_manager.output(out_n,out_arr,out_len);
+		for(int i=0;i<out_n;i++)
+		{
+			out_delay_buf[i]=0;
+
+			inner_stat.output_packet_num++;
+			inner_stat.output_packet_size+=out_len[i];
+		}
+
+
+	}
+
+	mylog(log_trace,"from_fec_to_normal input_len=%d,output_n=%d,input_seq=%u\n",len,out_n,read_u32(data));
+
+
+//	printf("<n:%d>",n);
+	/*
+	for(int i=0;i<n;i++)
+	{
+		delay_send(0,dest,s_arr[i],len_arr[i]);
+		//s_arr[i][len_arr[i]]=0;
+		//printf("<%s>\n",s_arr[i]);
+	}*/
+	//my_send(dest,data,len);
+	return 0;
+}
+
+
+int delay_send(my_time_t delay,const dest_t &dest,char *data,int len)
+{
+	//int rand=random()%100;
+	//mylog(log_info,"rand = %d\n",rand);
+
+	if (dest.cook&&random_drop != 0) {
+		if (get_true_random_number() % 10000 < (u32_t) random_drop) {
+			return 0;
+		}
+	}
+	return delay_manager.add(delay,dest,data,len);;
+}
+
+int print_parameter()
+{
+	mylog(log_info,"jitter_min=%d jitter_max=%d output_interval_min=%d output_interval_max=%d fec_timeout=%d fec_data_num=%d fec_redundant_num=%d fec_mtu=%d fec_queue_len=%d fec_mode=%d\n",
+			jitter_min/1000,jitter_max/1000,output_interval_min/1000,output_interval_max/1000,g_fec_timeout/1000,
+			g_fec_data_num,g_fec_redundant_num,g_fec_mtu,g_fec_queue_len,g_fec_mode);
+	return 0;
+}
+int handle_command(char *s)
+{
+	int len=strlen(s);
+	while(len>=1&&s[len-1]=='\n')
+		s[len-1]=0;
+	mylog(log_info,"got data from fifo,len=%d,s=[%s]\n",len,s);
+	int a=-1,b=-1;
+	if(strncmp(s,"fec",strlen("fec"))==0)
+	{
+		mylog(log_info,"got command [fec]\n");
+		sscanf(s,"fec %d:%d",&a,&b);
+		if(a<1||b<0||a+b>254)
+		{
+			mylog(log_warn,"invaild value\n");
+			return -1;
+		}
+		g_fec_data_num=a;
+		g_fec_redundant_num=b;
+	}
+	else if(strncmp(s,"mtu",strlen("mtu"))==0)
+	{
+		mylog(log_info,"got command [mtu]\n");
+		sscanf(s,"mtu %d",&a);
+		if(a<100||a>2000)
+		{
+			mylog(log_warn,"invaild value\n");
+			return -1;
+		}
+		g_fec_mtu=a;
+	}
+	else if(strncmp(s,"queue-len",strlen("queue-len"))==0)
+	{
+		mylog(log_info,"got command [queue-len]\n");
+		sscanf(s,"queue-len %d",&a);
+		if(a<1||a>10000)
+		{
+			mylog(log_warn,"invaild value\n");
+			return -1;
+		}
+		g_fec_queue_len=a;
+	}
+	else if(strncmp(s,"mode",strlen("mode"))==0)
+	{
+		mylog(log_info,"got command [mode]\n");
+		sscanf(s,"mode %d",&a);
+		if(a!=0&&a!=1)
+		{
+			mylog(log_warn,"invaild value\n");
+			return -1;
+		}
+		g_fec_mode=a;
+	}
+	else if(strncmp(s,"timeout",strlen("timeout"))==0)
+	{
+		mylog(log_info,"got command [timeout]\n");
+		sscanf(s,"timeout %d",&a);
+		if(a<0||a>1000)
+		{
+			mylog(log_warn,"invaild value\n");
+			return -1;
+		}
+		g_fec_timeout=a*1000;
+	}
+	else
+	{
+		mylog(log_info,"unknown command\n");
+	}
+	print_parameter();
+
+	return 0;
+}
+
+int unit_test()
+{
+	int i,j,k;
+	void *code=fec_new(3,6);
+	char arr[6][100]=
+	{
+		"aaa","bbb","ccc"
+		,"ddd","eee","fff"
+	};
+	char *data[6];
+	for(i=0;i<6;i++)
+	{
+		data[i]=arr[i];
+	}
+	rs_encode2(3,6,data,3);
+	//printf("%d %d",(int)(unsigned char)arr[5][0],(int)('a'^'b'^'c'^'d'^'e'));
+
+	for(i=0;i<6;i++)
+	{
+		printf("<%s>",data[i]);
+	}
+
+	data[0]=0;
+	//data[1]=0;
+	//data[5]=0;
+
+	int ret=rs_decode2(3,6,data,3);
+	printf("ret:%d\n",ret);
+
+	for(i=0;i<6;i++)
+	{
+		printf("<%s>",data[i]);
+	}
+	fec_free(code);
+
+
+	char arr2[6][100]=
+	{
+		"aaa11111","","ccc333333333"
+		,"ddd444","eee5555","ff6666"
+	};
+	blob_encode_t blob_encode;
+	for(int i=0;i<6;i++)
+		blob_encode.input(arr2[i],strlen(arr2[i]));
+
+	char **output;
+	int shard_len;
+	blob_encode.output(7,output,shard_len);
+
+
+	printf("<shard_len:%d>",shard_len);
+	blob_decode_t blob_decode;
+	for(int i=0;i<7;i++)
+	{
+		blob_decode.input(output[i],shard_len);
+	}
+
+	char **decode_output;
+	int * len_arr;
+	int num;
+
+
+	ret=blob_decode.output(num,decode_output,len_arr);
+
+	printf("<num:%d,ret:%d>\n",num,ret);
+	for(int i=0;i<num;i++)
+	{
+		char buf[1000]={0};
+		memcpy(buf,decode_output[i],len_arr[i]);
+		printf("<%d:%s>",len_arr[i],buf);
+	}
+	printf("\n");
+	static fec_encode_manager_t fec_encode_manager;
+	static fec_decode_manager_t fec_decode_manager;
+
+	{
+
+		string a = "11111";
+		string b = "22";
+		string c = "33333333";
+
+		fec_encode_manager.input((char *) a.c_str(), a.length());
+		fec_encode_manager.input((char *) b.c_str(), b.length());
+		fec_encode_manager.input((char *) c.c_str(), c.length());
+		fec_encode_manager.input(0, 0);
+
+		int n;
+		char **s_arr;
+		int *len;
+
+
+		fec_encode_manager.output(n,s_arr,len);
+		printf("<n:%d,len:%d>",n,len[0]);
+
+		for(int i=0;i<n;i++)
+		{
+			fec_decode_manager.input(s_arr[i],len[i]);
+		}
+
+		{
+			int n;char ** s_arr;int* len_arr;
+			fec_decode_manager.output(n,s_arr,len_arr);
+			printf("<n:%d>",n);
+			for(int i=0;i<n;i++)
+			{
+				s_arr[i][len_arr[i]]=0;
+				printf("<%s>\n",s_arr[i]);
+			}
+		}
+
+
+	}
+
+	{
+		string a = "aaaaaaa";
+		string b = "bbbbbbbbbbbbb";
+		string c = "ccc";
+
+
+		fec_encode_manager.input((char *) a.c_str(), a.length());
+		fec_encode_manager.input((char *) b.c_str(), b.length());
+		fec_encode_manager.input((char *) c.c_str(), c.length());
+		fec_encode_manager.input(0, 0);
+
+		int n;
+		char **s_arr;
+		int * len;
+
+
+		fec_encode_manager.output(n,s_arr,len);
+		printf("<n:%d,len:%d>",n,len[0]);
+
+		for(int i=0;i<n;i++)
+		{
+			if(i==1||i==3||i==5||i==0)
+			fec_decode_manager.input(s_arr[i],len[i]);
+		}
+
+		{
+			int n;char ** s_arr;int* len_arr;
+			fec_decode_manager.output(n,s_arr,len_arr);
+			printf("<n:%d>",n);
+			for(int i=0;i<n;i++)
+			{
+				s_arr[i][len_arr[i]]=0;
+				printf("<%s>\n",s_arr[i]);
+			}
+		}
+	}
+
+	for(int i=0;i<10;i++)
+	{
+		string a = "aaaaaaaaaaaaaaaaaaaaaaa";
+		string b = "bbbbbbbbbbbbb";
+		string c = "cccccccccccccccccc";
+
+
+
+		printf("======\n");
+		int n;
+		char **s_arr;
+		int * len;
+		fec_decode_manager.output(n,s_arr,len);
+
+		fec_encode_manager.reset_fec_parameter(3,2,g_fec_mtu,g_fec_queue_len,g_fec_timeout,1);
+
+		fec_encode_manager.input((char *) a.c_str(), a.length());
+		fec_encode_manager.output(n,s_arr,len);
+		assert(n==1);
+
+		fec_decode_manager.input(s_arr[0],len[0]);
+
+		fec_decode_manager.output(n,s_arr,len);
+		assert(n==1);
+		printf("%s\n",s_arr[0]);
+
+		fec_encode_manager.input((char *) b.c_str(), b.length());
+		fec_encode_manager.output(n,s_arr,len);
+		assert(n==1);
+		//fec_decode_manager.input(s_arr[0],len[0]);
+
+
+		fec_encode_manager.input((char *) c.c_str(), c.length());
+		fec_encode_manager.output(n,s_arr,len);
+
+		assert(n==3);
+
+		fec_decode_manager.input(s_arr[0],len[0]);
+		//printf("n=%d\n",n);
+
+
+		{
+			int n;
+			char **s_arr;
+			int * len;
+			fec_decode_manager.output(n,s_arr,len);
+			assert(n==1);
+			printf("%s\n",s_arr[0]);
+		}
+
+		fec_decode_manager.input(s_arr[1],len[1]);
+
+		{
+			int n;
+			char **s_arr;
+			int * len;
+			fec_decode_manager.output(n,s_arr,len);
+			assert(n==1);
+			printf("n=%d\n",n);
+			s_arr[0][len[0]]=0;
+			printf("%s\n",s_arr[0]);
+		}
+
+	}
+
+	return 0;
+}
+
+
+void process_arg(int argc, char *argv[])
+{
+	int is_client=0,is_server=0;
+	int i, j, k;
+	int opt;
+    static struct option long_options[] =
+      {
+		{"log-level", required_argument,    0, 1},
+		{"log-position", no_argument,    0, 1},
+		{"disable-color", no_argument,    0, 1},
+		{"disable-filter", no_argument,    0, 1},
+		{"disable-fec", no_argument,    0, 1},
+		{"disable-obscure", no_argument,    0, 1},
+		{"disable-xor", no_argument,    0, 1},
+		{"fix-latency", no_argument,    0, 1},
+		{"sock-buf", required_argument,    0, 1},
+		{"random-drop", required_argument,    0, 1},
+		{"report", required_argument,    0, 1},
+		{"delay-capacity", required_argument,    0, 1},
+		{"mtu", required_argument,    0, 1},
+		{"mode", required_argument,   0,1},
+		{"timeout", required_argument,   0,1},
+		{"decode-buf", required_argument,   0,1},
+		{"queue-len", required_argument,   0,'q'},
+		{"fec", required_argument,   0,'f'},
+		{"jitter", required_argument,   0,'j'},
+		{"fifo", required_argument,    0, 1},
+		{"sub-net", required_argument,    0, 1},
+		{"tun-dev", required_argument,    0, 1},
+		{"tun-mtu", required_argument,    0, 1},
+		{"disable-mssfix", no_argument,    0, 1},
+		{"keep-reconnect", no_argument,    0, 1},
+		{NULL, 0, 0, 0}
+      };
+    int option_index = 0;
+
+	for (i = 0; i < argc; i++)
+	{
+		if(strcmp(argv[i],"--unit-test")==0)
+		{
+			unit_test();
+			myexit(0);
+		}
+	}
+
+	for (i = 0; i < argc; i++)
+	{
+		if(strcmp(argv[i],"--log-level")==0)
+		{
+			if(i<argc -1)
+			{
+				sscanf(argv[i+1],"%d",&log_level);
+				if(0<=log_level&&log_level<log_end)
+				{
+				}
+				else
+				{
+					log_bare(log_fatal,"invalid log_level\n");
+					myexit(-1);
+				}
+			}
+		}
+		if(strcmp(argv[i],"--disable-color")==0)
+		{
+			enable_log_color=0;
+		}
+	}
+
+    mylog(log_info,"argc=%d ", argc);
+
+	for (i = 0; i < argc; i++) {
+		log_bare(log_info, "%s ", argv[i]);
+	}
+	log_bare(log_info, "\n");
+
+
+	int no_l = 1, no_r = 1;
+	while ((opt = getopt_long(argc, argv, "l:r:hcsk:j:f:p:n:i:q:",long_options,&option_index)) != -1)
+	{
+		//string opt_key;
+		//opt_key+=opt;
+		switch (opt)
+		{
+		case 'k':
+			sscanf(optarg,"%s\n",key_string);
+			mylog(log_debug,"key=%s\n",key_string);
+			if(strlen(key_string)==0)
+			{
+				mylog(log_fatal,"key len=0??\n");
+				myexit(-1);
+			}
+			break;
+		case 'j':
+			if (strchr(optarg, ':') == 0)
+			{
+				int jitter;
+				sscanf(optarg,"%d\n",&jitter);
+				if(jitter<0 ||jitter>1000*10)
+				{
+					mylog(log_fatal,"jitter must be between 0 and 10,000(10 second)\n");
+					myexit(-1);
+				}
+				jitter_min=0;
+				jitter_max=jitter;
+
+
+			}
+			else
+			{
+				sscanf(optarg,"%d:%d\n",&jitter_min,&jitter_max);
+				if(jitter_min<0 ||jitter_max<0||jitter_min>jitter_max)
+				{
+					mylog(log_fatal," must satisfy  0<=jmin<=jmax\n");
+					myexit(-1);
+				}
+			}
+			jitter_min*=1000;
+			jitter_max*=1000;
+			break;
+		case 'i':
+			if (strchr(optarg, ':') == 0)
+			{
+				int output_interval=-1;
+				sscanf(optarg,"%d\n",&output_interval);
+				if(output_interval<0||output_interval>1000*10)
+				{
+					mylog(log_fatal,"output_interval must be between 0 and 10,000(10 second)\n");
+					myexit(-1);
+				}
+				output_interval_min=output_interval_max=output_interval;
+			}
+			else
+			{
+				sscanf(optarg,"%d:%d\n",&output_interval_min,&output_interval_max);
+				if(output_interval_min<0 ||output_interval_max<0||output_interval_min>output_interval_max)
+				{
+					mylog(log_fatal," must satisfy  0<=output_interval_min<=output_interval_max\n");
+					myexit(-1);
+				}
+			}
+			output_interval_min*=1000;
+			output_interval_max*=1000;
+			break;
+		case 'f':
+			if (strchr(optarg, ':') == 0)
+			{
+				mylog(log_fatal,"invalid format for f");
+				myexit(-1);
+			}
+			else
+			{
+				sscanf(optarg,"%d:%d\n",&g_fec_data_num,&g_fec_redundant_num);
+				if(g_fec_data_num<1 ||g_fec_redundant_num<0||g_fec_data_num+g_fec_redundant_num>254)
+				{
+					mylog(log_fatal,"fec_data_num<1 ||fec_redundant_num<0||fec_data_num+fec_redundant_num>254\n");
+					myexit(-1);
+				}
+			}
+			break;
+		case 'q':
+			sscanf(optarg,"%d",&g_fec_queue_len);
+			if(g_fec_queue_len<1||g_fec_queue_len>10000)
+			{
+
+					mylog(log_fatal,"fec_pending_num should be between 1 and 10000\n");
+					myexit(-1);
+			}
+			break;
+		case 'c':
+			is_client = 1;
+			break;
+		case 's':
+			is_server = 1;
+			break;
+		case 'l':
+			no_l = 0;
+			if (strchr(optarg, ':') != 0)
+			{
+				sscanf(optarg, "%[^:]:%d", local_ip, &local_port);
+			}
+			else
+			{
+				mylog(log_fatal," -r ip:port\n");
+				myexit(1);
+				strcpy(local_ip, "127.0.0.1");
+				sscanf(optarg, "%d", &local_port);
+			}
+			break;
+		case 'r':
+			no_r = 0;
+			if (strchr(optarg, ':') != 0)
+			{
+				//printf("in :\n");
+				//printf("%s\n",optarg);
+				sscanf(optarg, "%[^:]:%d", remote_ip, &remote_port);
+				//printf("%d\n",remote_port);
+			}
+			else
+			{
+				mylog(log_fatal," -r ip:port\n");
+				myexit(1);
+				strcpy(remote_ip, "127.0.0.1");
+				sscanf(optarg, "%d", &remote_port);
+			}
+			break;
+		case 'h':
+			break;
+		case 1:
+			if(strcmp(long_options[option_index].name,"log-level")==0)
+			{
+			}
+			else if(strcmp(long_options[option_index].name,"disable-filter")==0)
+			{
+				disable_replay_filter=1;
+				//enable_log_color=0;
+			}
+			else if(strcmp(long_options[option_index].name,"disable-color")==0)
+			{
+				//enable_log_color=0;
+			}
+			else if(strcmp(long_options[option_index].name,"disable-fec")==0)
+			{
+				disable_fec=1;
+			}
+			else if(strcmp(long_options[option_index].name,"disable-obscure")==0)
+			{
+				mylog(log_info,"obscure disabled\n");
+				disable_obscure=1;
+			}
+			else if(strcmp(long_options[option_index].name,"disable-xor")==0)
+			{
+				mylog(log_info,"xor disabled\n");
+				disable_xor=1;
+			}
+			else if(strcmp(long_options[option_index].name,"fix-latency")==0)
+			{
+				mylog(log_info,"fix-latency enabled\n");
+				fix_latency=1;
+			}
+
+			else if(strcmp(long_options[option_index].name,"log-position")==0)
+			{
+				enable_log_position=1;
+			}
+			else if(strcmp(long_options[option_index].name,"random-drop")==0)
+			{
+				sscanf(optarg,"%d",&random_drop);
+				if(random_drop<0||random_drop>10000)
+				{
+					mylog(log_fatal,"random_drop must be between 0 10000 \n");
+					myexit(-1);
+				}
+				mylog(log_info,"random_drop=%d\n",random_drop);
+			}
+			else if(strcmp(long_options[option_index].name,"delay-capacity")==0)
+			{
+				sscanf(optarg,"%d",&delay_capacity);
+
+				if(delay_capacity<0)
+				{
+					mylog(log_fatal,"delay_capacity must be >=0 \n");
+					myexit(-1);
+				}
+			}
+			else if(strcmp(long_options[option_index].name,"report")==0)
+			{
+				sscanf(optarg,"%d",&report_interval);
+
+				if(report_interval<=0)
+				{
+					mylog(log_fatal,"report_interval must be >0 \n");
+					myexit(-1);
+				}
+			}
+			else if(strcmp(long_options[option_index].name,"sock-buf")==0)
+			{
+				int tmp=-1;
+				sscanf(optarg,"%d",&tmp);
+				if(10<=tmp&&tmp<=10*1024)
+				{
+					socket_buf_size=tmp*1024;
+				}
+				else
+				{
+					mylog(log_fatal,"sock-buf value must be between 1 and 10240 (kbyte) \n");
+					myexit(-1);
+				}
+			}
+			else if(strcmp(long_options[option_index].name,"decode-buf")==0)
+			{
+				sscanf(optarg,"%d",&fec_buff_num);
+				if(fec_buff_num<300 || fec_buff_num>20000)
+				{
+					mylog(log_fatal,"decode-buf value must be between 300 and 20000 (kbyte) \n");
+					myexit(-1);
+				}
+				mylog(log_info,"decode-buf=%d\n",fec_buff_num);
+			}
+			else if(strcmp(long_options[option_index].name,"mode")==0)
+			{
+				sscanf(optarg,"%d",&g_fec_mode);
+				if(g_fec_mode!=0&&g_fec_mode!=1)
+				{
+					mylog(log_fatal,"mode should be 0 or 1\n");
+					myexit(-1);
+				}
+			}
+			else if(strcmp(long_options[option_index].name,"mtu")==0)
+			{
+				sscanf(optarg,"%d",&g_fec_mtu);
+				if(g_fec_mtu<100||g_fec_mtu>2000)
+				{
+					mylog(log_fatal,"fec_mtu should be between 100 and 2000\n");
+					myexit(-1);
+				}
+			}
+			else if(strcmp(long_options[option_index].name,"timeout")==0)
+			{
+				sscanf(optarg,"%d",&g_fec_timeout);
+				if(g_fec_timeout<0||g_fec_timeout>1000)
+				{
+
+						mylog(log_fatal,"fec_pending_time should be between 0 and 1000(1s)\n");
+						myexit(-1);
+				}
+				g_fec_timeout*=1000;
+			}
+			else if(strcmp(long_options[option_index].name,"fifo")==0)
+			{
+				sscanf(optarg,"%s",fifo_file);
+
+				mylog(log_info,"fifo_file =%s \n",fifo_file);
+			}
+			else if(strcmp(long_options[option_index].name,"keep-reconnect")==0)
+			{
+				keep_reconnect=1;
+				mylog(log_info,"keep_reconnect enabled\n");
+			}
+			else if(strcmp(long_options[option_index].name,"sub-net")==0)
+			{
+				sscanf(optarg,"%s",sub_net);
+				mylog(log_info,"sub_net %s\n",sub_net);
+
+			}
+			else if(strcmp(long_options[option_index].name,"tun-dev")==0)
+			{
+				sscanf(optarg,"%s",tun_dev);
+				mylog(log_info,"tun_dev=%s\n",tun_dev);
+
+				mylog(log_info,"running at tun-dev mode\n");
+				working_mode=tun_dev_mode;
+			}
+			else if(strcmp(long_options[option_index].name,"tun-mtu")==0)
+			{
+				sscanf(optarg,"%d",&tun_mtu);
+				mylog(log_warn,"changed tun_mtu,tun_mtu=%d\n",tun_mtu);
+			}
+			else if(strcmp(long_options[option_index].name,"disable-mssfix")==0)
+			{
+				mssfix=0;
+				mylog(log_warn,"mssfix disabled\n");
+			}
+			else
+			{
+				mylog(log_fatal,"unknown option\n");
+				myexit(-1);
+			}
+			break;
+		default:
+			mylog(log_fatal,"unknown option <%x>", opt);
+			myexit(-1);
+		}
+	}
+
+	if (is_client == 0 && is_server == 0)
+	{
+		mylog(log_fatal,"-s -c hasnt been set\n");
+		myexit(-1);
+	}
+	if (is_client == 1 && is_server == 1)
+	{
+		mylog(log_fatal,"-s -c cant be both set\n");
+		myexit(-1);
+	}
+	if(is_client==1)
+	{
+		client_or_server=client_mode;
+	}
+	else
+	{
+		client_or_server=server_mode;
+	}
+
+
+	if(working_mode==tunnel_mode)
+	{
+		if (no_l)
+			mylog(log_fatal,"error: -l not found\n");
+		if (no_r)
+			mylog(log_fatal,"error: -r not found\n");
+		if (no_l || no_r)
+			myexit(-1);
+	}
+	else if(working_mode==tun_dev_mode)
+	{
+		if(client_or_server==client_mode&&no_r)
+		{
+			mylog(log_fatal,"error: -r not found\n");
+			myexit(-1);
+		}
+		else if(client_or_server==server_mode&&no_l)
+		{
+			mylog(log_fatal,"error: -l not found\n");
+			myexit(-1);
+		}
+	}
+
+	print_parameter();
+
+}
+
--- a/misc.h
+++ b/misc.h
@@ -0,0 +1,73 @@
+/*
+ * misc.h
+ *
+ *  Created on: Oct 26, 2017
+ *      Author: root
+ */
+
+#ifndef MISC_H_
+#define MISC_H_
+
+#include "common.h"
+#include "connection.h"
+#include "fd_manager.h"
+#include "delay_manager.h"
+#include "fec_manager.h"
+
+
+
+extern char fifo_file[1000];
+
+extern int mtu_warn;
+
+extern int disable_mtu_warn;
+extern int disable_fec;
+
+extern int debug_force_flush_fec;
+
+
+extern int jitter_min;
+extern int jitter_max;
+
+extern int output_interval_min;
+extern int output_interval_max;
+
+extern int fix_latency;
+
+extern u32_t local_ip_uint32,remote_ip_uint32;
+extern char local_ip[100], remote_ip[100];
+extern int local_port, remote_port;
+
+extern conn_manager_t conn_manager;
+extern delay_manager_t delay_manager;
+extern fd_manager_t fd_manager;
+
+extern int time_mono_test;
+
+extern int delay_capacity;
+
+extern int keep_reconnect;
+
+extern int tun_mtu;
+
+extern int mssfix;
+
+
+int from_normal_to_fec(conn_info_t & conn_info,char *data,int len,int & out_n,char **&out_arr,int *&out_len,my_time_t *&out_delay);
+int from_fec_to_normal(conn_info_t & conn_info,char *data,int len,int & out_n,char **&out_arr,int *&out_len,my_time_t *&out_delay);
+
+int delay_send(my_time_t delay,const dest_t &dest,char *data,int len);
+int print_parameter();
+int handle_command(char *s);
+
+int unit_test();
+
+//void print_help();
+
+void process_arg(int argc, char *argv[]);
+
+extern char sub_net[100];
+extern u32_t sub_net_uint32;
+extern char tun_dev[100];
+
+#endif /* MISC_H_ */
--- a/packet.cpp
+++ b/packet.cpp
@@ -27,7 +27,7 @@ int random_drop=0;

 char key_string[1000]= "";

-int local_listen_fd=-1;
+//int local_listen_fd=-1;


 void encrypt_0(char * input,int &len,char *key)
@@ -162,10 +162,11 @@ int sendto_fd_ip_port (int fd,u32_t ip,int port,char * buf, int len,int flags)
 			(struct sockaddr *) &tmp_sockaddr,
 			sizeof(tmp_sockaddr));
 }
+/*
 int sendto_ip_port (u32_t ip,int port,char * buf, int len,int flags)
 {
 	return sendto_fd_ip_port(local_listen_fd,ip,port,buf,len,flags);
-}
+}*/

 int send_fd (int fd,char * buf, int len,int flags)
 {
@@ -176,23 +177,31 @@ int my_send(const dest_t &dest,char *data,int len)
 {
 	if(dest.cook)
 	{
-		put_crc32(data,len);
-		if(!disable_obscure)do_obscure(data,len);
-		if(!disable_xor)encrypt_0(data,len,key_string);
+		do_cook(data,len);
 	}
 	switch(dest.type)
 	{
-		case type_ip_port:
+		case type_fd_ip_port:
 		{
-			return sendto_ip_port(dest.inner.ip_port.ip,dest.inner.ip_port.port,data,len,0);
+			return sendto_fd_ip_port(dest.inner.fd,dest.inner.fd_ip_port.ip_port.ip,dest.inner.fd_ip_port.ip_port.port,data,len,0);
 			break;
 		}
-		case type_ip_port_conv:
+		case type_fd64_ip_port:
 		{
-			char *new_data;
-			int new_len;
-			put_conv(dest.conv,data,len,new_data,new_len);
-			return sendto_ip_port(dest.inner.ip_port.ip,dest.inner.ip_port.port,new_data,new_len,0);
+			if(!fd_manager.exist(dest.inner.fd64)) return -1;
+			int fd=fd_manager.to_fd(dest.inner.fd64);
+
+			return sendto_fd_ip_port(fd,dest.inner.fd64_ip_port.ip_port.ip,dest.inner.fd64_ip_port.ip_port.port,data,len,0);
+			break;
+		}
+		case type_fd:
+		{
+			return send_fd(dest.inner.fd,data,len,0);
+			break;
+		}
+		case type_write_fd:
+		{
+			return write(dest.inner.fd,data,len);
 			break;
 		}
 		case type_fd64:
@@ -200,9 +209,25 @@ int my_send(const dest_t &dest,char *data,int len)

 			if(!fd_manager.exist(dest.inner.fd64)) return -1;
 			int fd=fd_manager.to_fd(dest.inner.fd64);
+
 			return send_fd(fd,data,len,0);
 			break;
 		}
+		/*
+		case type_fd64_ip_port_conv:
+		{
+			if(!fd_manager.exist(dest.inner.fd64)) return -1;
+			int fd=fd_manager.to_fd(dest.inner.fd64);
+
+			char *new_data;
+			int new_len;
+
+			put_conv(dest.conv,data,len,new_data,new_len);
+			return sendto_fd_ip_port(fd,dest.inner.fd64_ip_port.ip_port.ip,dest.inner.fd64_ip_port.ip_port.port,new_data,new_len,0);
+			break;
+		}*/
+
+		/*
 		case type_fd64_conv:
 		{
 			char *new_data;
@@ -212,7 +237,7 @@ int my_send(const dest_t &dest,char *data,int len)
 			if(!fd_manager.exist(dest.inner.fd64)) return -1;
 			int fd=fd_manager.to_fd(dest.inner.fd64);
 			return send_fd(fd,new_data,new_len,0);
-		}
+		}*/
 		/*
 		case type_fd:
 		{
@@ -300,6 +325,14 @@ int put_crc32(char * s,int &len)
 	return 0;
 }

+int do_cook(char * data,int &len)
+{
+	put_crc32(data,len);
+	if(!disable_obscure)do_obscure(data,len);
+	if(!disable_xor)encrypt_0(data,len,key_string);
+	return 0;
+}
+
 int de_cook(char * s,int &len)
 {
 	if(!disable_xor)decrypt_0(s,len,key_string);
@@ -362,3 +395,6 @@ int get_conv(u32_t &conv,const char *input,int len_in,char *&output,int &len_out
 	}
 	return 0;
 }
+
+
+
--- a/packet.h
+++ b/packet.h
@@ -21,7 +21,6 @@ extern u64_t dup_packet_recv_count;
 extern char key_string[1000];
 extern int disable_replay_filter;
 extern int random_drop;
-extern int local_listen_fd;
 extern int disable_obscure;
 extern int disable_xor;

@@ -43,5 +42,6 @@ int put_conv(u32_t conv,const char * input,int len_in,char *&output,int &len_out
 int get_conv(u32_t &conv,const char *input,int len_in,char *&output,int &len_out );
 int put_crc32(char * s,int &len);
 int rm_crc32(char * s,int &len);
+int do_cook(char * data,int &len);
 int de_cook(char * s,int &len);
 #endif /* PACKET_H_ */
--- a/tunnel.cpp
+++ b/tunnel.cpp
@@ -0,0 +1,705 @@
+/*
+ * tunnel.cpp
+ *
+ *  Created on: Oct 26, 2017
+ *      Author: root
+ */
+
+#include "tunnel.h"
+
+
+int tunnel_client_event_loop()
+{
+	//char buf[buf_len];
+	int i, j, k;int ret;
+	int yes = 1;
+	int epoll_fd;
+	int remote_fd;
+	fd64_t remote_fd64;
+
+    conn_info_t *conn_info_p=new conn_info_t;
+    conn_info_t &conn_info=*conn_info_p;  //huge size of conn_info,do not allocate on stack
+    //conn_info.conv_manager.reserve();
+	//conn_info.fec_encode_manager.re_init(fec_data_num,fec_redundant_num,fec_mtu,fec_pending_num,fec_pending_time,fec_type);
+
+
+	int local_listen_fd;
+	//fd64_t local_listen_fd64;
+    new_listen_socket(local_listen_fd,local_ip_uint32,local_port);
+    //local_listen_fd64=fd_manager.create(local_listen_fd);
+
+	epoll_fd = epoll_create1(0);
+	assert(epoll_fd>0);
+
+	const int max_events = 4096;
+	struct epoll_event ev, events[max_events];
+	if (epoll_fd < 0) {
+		mylog(log_fatal,"epoll return %d\n", epoll_fd);
+		myexit(-1);
+	}
+
+	ev.events = EPOLLIN;
+	ev.data.u64 = local_listen_fd;
+	ret = epoll_ctl(epoll_fd, EPOLL_CTL_ADD, local_listen_fd, &ev);
+	if (ret!=0) {
+		mylog(log_fatal,"add  udp_listen_fd error\n");
+		myexit(-1);
+	}
+
+	assert(new_connected_socket(remote_fd,remote_ip_uint32,remote_port)==0);
+	remote_fd64=fd_manager.create(remote_fd);
+
+	mylog(log_debug,"remote_fd64=%llu\n",remote_fd64);
+
+	ev.events = EPOLLIN;
+	ev.data.u64 = remote_fd64;
+
+	ret = epoll_ctl(epoll_fd, EPOLL_CTL_ADD, remote_fd, &ev);
+	if (ret!= 0) {
+		mylog(log_fatal,"add raw_fd error\n");
+		myexit(-1);
+	}
+
+	ev.events = EPOLLIN;
+	ev.data.u64 = delay_manager.get_timer_fd();
+
+	mylog(log_debug,"delay_manager.get_timer_fd()=%d\n",delay_manager.get_timer_fd());
+	ret = epoll_ctl(epoll_fd, EPOLL_CTL_ADD, delay_manager.get_timer_fd(), &ev);
+	if (ret!= 0) {
+		mylog(log_fatal,"add delay_manager.get_timer_fd() error\n");
+		myexit(-1);
+	}
+
+	u64_t tmp_fd64=conn_info.fec_encode_manager.get_timer_fd64();
+	ev.events = EPOLLIN;
+	ev.data.u64 = tmp_fd64;
+
+	mylog(log_debug,"conn_info.fec_encode_manager.get_timer_fd64()=%llu\n",conn_info.fec_encode_manager.get_timer_fd64());
+	ret = epoll_ctl(epoll_fd, EPOLL_CTL_ADD, fd_manager.to_fd(tmp_fd64), &ev);
+	if (ret!= 0) {
+		mylog(log_fatal,"add fec_encode_manager.get_timer_fd64() error\n");
+		myexit(-1);
+	}
+
+	//my_timer_t timer;
+	conn_info.timer.add_fd_to_epoll(epoll_fd);
+	conn_info.timer.set_timer_repeat_us(timer_interval*1000);
+
+	mylog(log_debug,"conn_info.timer.get_timer_fd()=%d\n",conn_info.timer.get_timer_fd());
+
+
+
+	int fifo_fd=-1;
+
+	if(fifo_file[0]!=0)
+	{
+		fifo_fd=create_fifo(fifo_file);
+		ev.events = EPOLLIN;
+		ev.data.u64 = fifo_fd;
+
+		ret = epoll_ctl(epoll_fd, EPOLL_CTL_ADD, fifo_fd, &ev);
+		if (ret!= 0) {
+			mylog(log_fatal,"add fifo_fd to epoll error %s\n",strerror(errno));
+			myexit(-1);
+		}
+		mylog(log_info,"fifo_file=%s\n",fifo_file);
+	}
+
+	while(1)////////////////////////
+	{
+		if(about_to_exit) myexit(0);
+
+		int nfds = epoll_wait(epoll_fd, events, max_events, 180 * 1000);
+		if (nfds < 0) {  //allow zero
+			if(errno==EINTR  )
+			{
+				mylog(log_info,"epoll interrupted by signal\n");
+				myexit(0);
+			}
+			else
+			{
+				mylog(log_fatal,"epoll_wait return %d\n", nfds);
+				myexit(-1);
+			}
+		}
+		int idx;
+		for (idx = 0; idx < nfds; ++idx) {
+			if(events[idx].data.u64==(u64_t)conn_info.timer.get_timer_fd())
+			{
+				uint64_t value;
+				read(conn_info.timer.get_timer_fd(), &value, 8);
+				conn_info.conv_manager.clear_inactive();
+				mylog(log_trace,"events[idx].data.u64==(u64_t)conn_info.timer.get_timer_fd()\n");
+
+				conn_info.stat.report_as_client();
+
+				if(debug_force_flush_fec)
+				{
+				int  out_n;char **out_arr;int *out_len;my_time_t *out_delay;
+				dest_t dest;
+				dest.type=type_fd64;
+				dest.inner.fd64=remote_fd64;
+				dest.cook=1;
+				from_normal_to_fec(conn_info,0,0,out_n,out_arr,out_len,out_delay);
+				for(int i=0;i<out_n;i++)
+				{
+					delay_send(out_delay[i],dest,out_arr[i],out_len[i]);
+				}
+				}
+			}
+			else if (events[idx].data.u64 == (u64_t)fifo_fd)
+			{
+				char buf[buf_len];
+				int len=read (fifo_fd, buf, sizeof (buf));
+				if(len<0)
+				{
+					mylog(log_warn,"fifo read failed len=%d,errno=%s\n",len,strerror(errno));
+					continue;
+				}
+				buf[len]=0;
+				handle_command(buf);
+			}
+			else if (events[idx].data.u64 == (u64_t)local_listen_fd||events[idx].data.u64 == conn_info.fec_encode_manager.get_timer_fd64())
+			{
+				char data[buf_len];
+				int data_len;
+				ip_port_t ip_port;
+				u32_t conv;
+				int  out_n;char **out_arr;int *out_len;my_time_t *out_delay;
+				dest_t dest;
+				dest.type=type_fd64;
+				dest.inner.fd64=remote_fd64;
+				dest.cook=1;
+
+				if(events[idx].data.u64 == conn_info.fec_encode_manager.get_timer_fd64())
+				{
+					fd64_t fd64=events[idx].data.u64;
+					mylog(log_trace,"events[idx].data.u64 == conn_info.fec_encode_manager.get_timer_fd64()\n");
+
+					//mylog(log_info,"timer!!!\n");
+					uint64_t value;
+					if(!fd_manager.exist(fd64))   //fd64 has been closed
+					{
+						mylog(log_trace,"!fd_manager.exist(fd64)");
+						continue;
+					}
+					if((ret=read(fd_manager.to_fd(fd64), &value, 8))!=8)
+					{
+						mylog(log_trace,"(ret=read(fd_manager.to_fd(fd64), &value, 8))!=8,ret=%d\n",ret);
+						continue;
+					}
+					if(value==0)
+					{
+						mylog(log_debug,"value==0\n");
+						continue;
+					}
+					assert(value==1);
+					from_normal_to_fec(conn_info,0,0,out_n,out_arr,out_len,out_delay);
+					//from_normal_to_fec(conn_info,0,0,out_n,out_arr,out_len,out_delay);
+				}
+				else//events[idx].data.u64 == (u64_t)local_listen_fd
+				{
+					mylog(log_trace,"events[idx].data.u64 == (u64_t)local_listen_fd\n");
+					struct sockaddr_in udp_new_addr_in={0};
+					socklen_t udp_new_addr_len = sizeof(sockaddr_in);
+					if ((data_len = recvfrom(local_listen_fd, data, max_data_len, 0,
+							(struct sockaddr *) &udp_new_addr_in, &udp_new_addr_len)) == -1) {
+						mylog(log_error,"recv_from error,this shouldnt happen,err=%s,but we can try to continue\n",strerror(errno));
+						continue;
+						//mylog(log_error,"recv_from error,this shouldnt happen at client\n");
+						//myexit(1);
+					};
+
+					if(!disable_mtu_warn&&data_len>=mtu_warn)
+					{
+						mylog(log_warn,"huge packet,data len=%d (>=%d).strongly suggested to set a smaller mtu at upper level,to get rid of this warn\n ",data_len,mtu_warn);
+					}
+					mylog(log_trace,"Received packet from %s:%d,len: %d\n", inet_ntoa(udp_new_addr_in.sin_addr),
+							ntohs(udp_new_addr_in.sin_port),data_len);
+
+					ip_port.ip=udp_new_addr_in.sin_addr.s_addr;
+					ip_port.port=ntohs(udp_new_addr_in.sin_port);
+
+					u64_t u64=ip_port.to_u64();
+
+					if(!conn_info.conv_manager.is_u64_used(u64))
+					{
+						if(conn_info.conv_manager.get_size() >=max_conv_num)
+						{
+							mylog(log_warn,"ignored new udp connect bc max_conv_num exceed\n");
+							continue;
+						}
+						conv=conn_info.conv_manager.get_new_conv();
+						conn_info.conv_manager.insert_conv(conv,u64);
+						mylog(log_info,"new packet from %s:%d,conv_id=%x\n",inet_ntoa(udp_new_addr_in.sin_addr),ntohs(udp_new_addr_in.sin_port),conv);
+					}
+					else
+					{
+						conv=conn_info.conv_manager.find_conv_by_u64(u64);
+						mylog(log_trace,"conv=%d\n",conv);
+					}
+					conn_info.conv_manager.update_active_time(conv);
+					char * new_data;
+					int new_len;
+					put_conv(conv,data,data_len,new_data,new_len);
+
+
+					mylog(log_trace,"data_len=%d new_len=%d\n",data_len,new_len);
+					//dest.conv=conv;
+					from_normal_to_fec(conn_info,new_data,new_len,out_n,out_arr,out_len,out_delay);
+
+				}
+				mylog(log_trace,"out_n=%d\n",out_n);
+				for(int i=0;i<out_n;i++)
+				{
+					delay_send(out_delay[i],dest,out_arr[i],out_len[i]);
+				}
+				//my_send(dest,data,data_len);
+			}
+		    else if (events[idx].data.u64 == (u64_t)delay_manager.get_timer_fd()) {
+				uint64_t value;
+				read(delay_manager.get_timer_fd(), &value, 8);
+				mylog(log_trace,"events[idx].data.u64 == (u64_t)delay_manager.get_timer_fd()\n");
+				//printf("<timerfd_triggered, %d>",delay_mp.size());
+				//fflush(stdout);
+			}
+			else if(events[idx].data.u64>u32_t(-1) )
+			{
+				char data[buf_len];
+				if(!fd_manager.exist(events[idx].data.u64))   //fd64 has been closed
+				{
+					mylog(log_trace,"!fd_manager.exist(events[idx].data.u64)");
+					continue;
+				}
+				assert(events[idx].data.u64==remote_fd64);
+				int fd=fd_manager.to_fd(remote_fd64);
+				int data_len =recv(fd,data,max_data_len,0);
+				mylog(log_trace, "received data from udp fd %d, len=%d\n", remote_fd,data_len);
+				if(data_len<0)
+				{
+					if(errno==ECONNREFUSED)
+					{
+						//conn_manager.clear_list.push_back(udp_fd);
+						mylog(log_debug, "recv failed %d ,udp_fd%d,errno:%s\n", data_len,remote_fd,strerror(errno));
+					}
+
+					mylog(log_warn, "recv failed %d ,udp_fd%d,errno:%s\n", data_len,remote_fd,strerror(errno));
+					continue;
+				}
+				if(!disable_mtu_warn&&data_len>mtu_warn)
+				{
+					mylog(log_warn,"huge packet,data len=%d (>%d).strongly suggested to set a smaller mtu at upper level,to get rid of this warn\n ",data_len,mtu_warn);
+				}
+
+				if(de_cook(data,data_len)!=0)
+				{
+					mylog(log_debug,"de_cook error");
+					continue;
+				}
+
+				int  out_n;char **out_arr;int *out_len;my_time_t *out_delay;
+				from_fec_to_normal(conn_info,data,data_len,out_n,out_arr,out_len,out_delay);
+
+				mylog(log_trace,"out_n=%d\n",out_n);
+
+				for(int i=0;i<out_n;i++)
+				{
+					u32_t conv;
+					char *new_data;
+					int new_len;
+					if(get_conv(conv,out_arr[i],out_len[i],new_data,new_len)!=0)
+					{
+						mylog(log_debug,"get_conv(conv,out_arr[i],out_len[i],new_data,new_len)!=0");
+						continue;
+					}
+					if(!conn_info.conv_manager.is_conv_used(conv))
+					{
+						mylog(log_trace,"!conn_info.conv_manager.is_conv_used(conv)");
+						continue;
+					}
+
+					conn_info.conv_manager.update_active_time(conv);
+
+					u64_t u64=conn_info.conv_manager.find_u64_by_conv(conv);
+					dest_t dest;
+					dest.inner.fd_ip_port.fd=local_listen_fd;
+					dest.inner.fd_ip_port.ip_port.from_u64(u64);
+					dest.type=type_fd_ip_port;
+					//dest.conv=conv;
+
+					delay_send(out_delay[i],dest,new_data,new_len);
+				}
+				//mylog(log_trace,"[%s] send packet\n",dest.inner.ip_port.to_s());
+			}
+			else
+			{
+				mylog(log_fatal,"unknown fd,this should never happen\n");
+				myexit(-1);
+			}
+		}
+		delay_manager.check();
+	}
+	return 0;
+}
+
+int tunnel_server_event_loop()
+{
+	//char buf[buf_len];
+	int i, j, k;int ret;
+	int yes = 1;
+	int epoll_fd;
+	int remote_fd;
+
+//    conn_info_t conn_info;
+	int local_listen_fd;
+//	fd64_t local_listen_fd64;
+    new_listen_socket(local_listen_fd,local_ip_uint32,local_port);
+   // local_listen_fd64=fd_manager.create(local_listen_fd);
+
+	epoll_fd = epoll_create1(0);
+	assert(epoll_fd>0);
+
+	const int max_events = 4096;
+	struct epoll_event ev, events[max_events];
+	if (epoll_fd < 0) {
+		mylog(log_fatal,"epoll return %d\n", epoll_fd);
+		myexit(-1);
+	}
+
+	ev.events = EPOLLIN;
+	ev.data.u64 = local_listen_fd;
+	ret = epoll_ctl(epoll_fd, EPOLL_CTL_ADD, local_listen_fd, &ev);
+	if (ret!=0) {
+		mylog(log_fatal,"add  udp_listen_fd error\n");
+		myexit(-1);
+	}
+
+	ev.events = EPOLLIN;
+	ev.data.u64 = delay_manager.get_timer_fd();
+	ret = epoll_ctl(epoll_fd, EPOLL_CTL_ADD, delay_manager.get_timer_fd(), &ev);
+	if (ret!= 0) {
+		mylog(log_fatal,"add delay_manager.get_timer_fd() error\n");
+		myexit(-1);
+	}
+
+	mylog(log_debug," delay_manager.get_timer_fd() =%d\n", delay_manager.get_timer_fd());
+
+	mylog(log_info,"now listening at %s:%d\n",my_ntoa(local_ip_uint32),local_port);
+
+	my_timer_t timer;
+	timer.add_fd_to_epoll(epoll_fd);
+	timer.set_timer_repeat_us(timer_interval*1000);
+
+	mylog(log_debug," timer.get_timer_fd() =%d\n",timer.get_timer_fd());
+
+	int fifo_fd=-1;
+
+	if(fifo_file[0]!=0)
+	{
+		fifo_fd=create_fifo(fifo_file);
+		ev.events = EPOLLIN;
+		ev.data.u64 = fifo_fd;
+
+		ret = epoll_ctl(epoll_fd, EPOLL_CTL_ADD, fifo_fd, &ev);
+		if (ret!= 0) {
+			mylog(log_fatal,"add fifo_fd to epoll error %s\n",strerror(errno));
+			myexit(-1);
+		}
+		mylog(log_info,"fifo_file=%s\n",fifo_file);
+	}
+
+	while(1)////////////////////////
+	{
+
+		if(about_to_exit) myexit(0);
+
+		int nfds = epoll_wait(epoll_fd, events, max_events, 180 * 1000);
+		if (nfds < 0) {  //allow zero
+			if(errno==EINTR  )
+			{
+				mylog(log_info,"epoll interrupted by signal\n");
+				myexit(0);
+			}
+			else
+			{
+				mylog(log_fatal,"epoll_wait return %d\n", nfds);
+				myexit(-1);
+			}
+		}
+		int idx;
+		for (idx = 0; idx < nfds; ++idx)
+		{
+			/*
+			if ((events[idx].data.u64 ) == (u64_t)timer_fd)
+			{
+				conn_manager.clear_inactive();
+				u64_t dummy;
+				read(timer_fd, &dummy, 8);
+				//current_time_rough=get_current_time();
+			}
+			else */
+			if(events[idx].data.u64==(u64_t)timer.get_timer_fd())
+			{
+				uint64_t value;
+				read(timer.get_timer_fd(), &value, 8);
+				conn_manager.clear_inactive();
+				mylog(log_trace,"events[idx].data.u64==(u64_t)timer.get_timer_fd()\n");
+				//conn_info.conv_manager.clear_inactive();
+			}
+
+			else if (events[idx].data.u64 == (u64_t)fifo_fd)
+			{
+				char buf[buf_len];
+				int len=read (fifo_fd, buf, sizeof (buf));
+				if(len<0)
+				{
+					mylog(log_warn,"fifo read failed len=%d,errno=%s\n",len,strerror(errno));
+					continue;
+				}
+				buf[len]=0;
+				handle_command(buf);
+			}
+
+			else if (events[idx].data.u64 == (u64_t)local_listen_fd)
+			{
+
+				mylog(log_trace,"events[idx].data.u64 == (u64_t)local_listen_fd\n");
+				//int recv_len;
+				char data[buf_len];
+				int data_len;
+				struct sockaddr_in udp_new_addr_in={0};
+				socklen_t udp_new_addr_len = sizeof(sockaddr_in);
+				if ((data_len = recvfrom(local_listen_fd, data, max_data_len, 0,
+						(struct sockaddr *) &udp_new_addr_in, &udp_new_addr_len)) == -1) {
+					mylog(log_error,"recv_from error,this shouldnt happen,err=%s,but we can try to continue\n",strerror(errno));
+					continue;
+					//myexit(1);
+				};
+				mylog(log_trace,"Received packet from %s:%d,len: %d\n", inet_ntoa(udp_new_addr_in.sin_addr),
+						ntohs(udp_new_addr_in.sin_port),data_len);
+
+				if(!disable_mtu_warn&&data_len>=mtu_warn)///////////////////////delete this for type 0 in furture
+				{
+					mylog(log_warn,"huge packet,data len=%d (>=%d).strongly suggested to set a smaller mtu at upper level,to get rid of this warn\n ",data_len,mtu_warn);
+				}
+
+
+				if(de_cook(data,data_len)!=0)
+				{
+					mylog(log_debug,"de_cook error");
+					continue;
+				}
+
+
+				ip_port_t ip_port;
+				ip_port.ip=udp_new_addr_in.sin_addr.s_addr;
+				ip_port.port=ntohs(udp_new_addr_in.sin_port);
+				mylog(log_trace,"ip_port= %s\n",ip_port.to_s());
+				if(!conn_manager.exist(ip_port))
+				{
+					if(conn_manager.mp.size() >=max_conn_num)
+					{
+						mylog(log_warn,"new connection %s ignored bc max_conn_num exceed\n",ip_port.to_s());
+						continue;
+					}
+
+					conn_manager.insert(ip_port);
+					conn_info_t &conn_info=conn_manager.find(ip_port);
+					//conn_info.fec_encode_manager.re_init(fec_data_num,fec_redundant_num,fec_mtu,fec_pending_num,fec_pending_time,fec_type);
+					//conn_info.conv_manager.reserve();  //already reserved in constructor
+
+					u64_t fec_fd64=conn_info.fec_encode_manager.get_timer_fd64();
+					mylog(log_debug,"fec_fd64=%llu\n",fec_fd64);
+					ev.events = EPOLLIN;
+					ev.data.u64 = fec_fd64;
+					ret = epoll_ctl(epoll_fd, EPOLL_CTL_ADD, fd_manager.to_fd(fec_fd64), &ev);
+
+					fd_manager.get_info(fec_fd64).ip_port=ip_port;
+
+					conn_info.timer.add_fd64_to_epoll(epoll_fd);
+					conn_info.timer.set_timer_repeat_us(timer_interval*1000);
+
+					mylog(log_debug,"conn_info.timer.get_timer_fd64()=%llu\n",conn_info.timer.get_timer_fd64());
+
+					u64_t timer_fd64=conn_info.timer.get_timer_fd64();
+					fd_manager.get_info(timer_fd64).ip_port=ip_port;
+
+					mylog(log_info,"new connection from %s\n",ip_port.to_s());
+
+				}
+				conn_info_t &conn_info=conn_manager.find(ip_port);
+
+				conn_info.update_active_time();
+				int  out_n;char **out_arr;int *out_len;my_time_t *out_delay;
+				from_fec_to_normal(conn_info,data,data_len,out_n,out_arr,out_len,out_delay);
+
+				mylog(log_trace,"out_n= %d\n",out_n);
+				for(int i=0;i<out_n;i++)
+				{
+					u32_t conv;
+					char *new_data;
+					int new_len;
+					if(get_conv(conv,out_arr[i],out_len[i],new_data,new_len)!=0)
+					{
+						mylog(log_debug,"get_conv failed");
+						continue;
+					}
+
+					/*
+					id_t tmp_conv_id;
+					memcpy(&tmp_conv_id,&data_[0],sizeof(tmp_conv_id));
+					tmp_conv_id=ntohl(tmp_conv_id);*/
+
+					if (!conn_info.conv_manager.is_conv_used(conv))
+					{
+						if(conn_info.conv_manager.get_size() >=max_conv_num)
+						{
+							mylog(log_warn,"ignored new udp connect bc max_conv_num exceed\n");
+							continue;
+						}
+
+						int new_udp_fd;
+						ret=new_connected_socket(new_udp_fd,remote_ip_uint32,remote_port);
+
+						if (ret != 0) {
+							mylog(log_warn, "[%s]new_connected_socket failed\n",ip_port.to_s());
+							continue;
+						}
+
+						fd64_t fd64 = fd_manager.create(new_udp_fd);
+						ev.events = EPOLLIN;
+						ev.data.u64 = fd64;
+						ret = epoll_ctl(epoll_fd, EPOLL_CTL_ADD, new_udp_fd, &ev);
+
+						conn_info.conv_manager.insert_conv(conv, fd64);
+						fd_manager.get_info(fd64).ip_port=ip_port;
+
+
+						mylog(log_info,"[%s]new conv %x,fd %d created,fd64=%llu\n",ip_port.to_s(),conv,new_udp_fd,fd64);
+						//assert(!conn_manager.exist_fd64(fd64));
+
+						//conn_manager.insert_fd64(fd64,ip_port);
+					}
+					conn_info.conv_manager.update_active_time(conv);
+					fd64_t fd64= conn_info.conv_manager.find_u64_by_conv(conv);
+					//int fd=fd_manager.fd64_to_fd(fd64);
+					dest_t dest;
+					dest.type=type_fd64;
+					dest.inner.fd64=fd64;
+					//dest.conv=conv;
+					delay_send(out_delay[i],dest,new_data,new_len);
+				}
+			}
+		    else if (events[idx].data.u64 == (u64_t)delay_manager.get_timer_fd()) {
+				uint64_t value;
+				read(delay_manager.get_timer_fd(), &value, 8);
+				mylog(log_trace,"events[idx].data.u64 == (u64_t)delay_manager.get_timer_fd()\n");
+			}
+			else if (events[idx].data.u64 >u32_t(-1))
+			{
+				char data[buf_len];
+				int data_len;
+				u32_t conv;
+				fd64_t fd64=events[idx].data.u64;
+				mylog(log_trace,"events[idx].data.u64 >u32_t(-1),%llu\n",(u64_t)events[idx].data.u64);
+				if(!fd_manager.exist(fd64))   //fd64 has been closed
+				{
+					mylog(log_trace,"!fd_manager.exist(fd64)\n");
+					continue;
+				}
+
+				assert(fd_manager.exist_info(fd64));
+				ip_port_t ip_port=fd_manager.get_info(fd64).ip_port;
+				assert(conn_manager.exist(ip_port));
+
+				conn_info_t &conn_info=conn_manager.find(ip_port);
+				//conn_info.update_active_time(); //cant put it here
+
+				int  out_n=-2;char **out_arr;int *out_len;my_time_t *out_delay;
+
+				dest_t dest;
+				dest.inner.fd_ip_port.fd=local_listen_fd;
+				dest.inner.fd_ip_port.ip_port=ip_port;
+				dest.type=type_fd_ip_port;
+				dest.cook=1;
+
+				if(fd64==conn_info.fec_encode_manager.get_timer_fd64())
+				{
+					//mylog(log_infol,"timer!!!\n");
+					uint64_t value;
+					if((ret=read(fd_manager.to_fd(fd64), &value, 8))!=8)
+					{
+						mylog(log_trace,"fd_manager.to_fd(fd64), &value, 8)!=8 ,%d\n",ret);
+						continue;
+					}
+					if(value==0)
+					{
+						mylog(log_trace,"value==0\n");
+						continue;
+					}
+					assert(value==1);
+					from_normal_to_fec(conn_info,0,0,out_n,out_arr,out_len,out_delay);
+				}
+				else if(fd64==conn_info.timer.get_timer_fd64())
+				{
+					uint64_t value;
+					read(conn_info.timer.get_timer_fd(), &value, 8);
+					conn_info.conv_manager.clear_inactive();
+					if(debug_force_flush_fec)
+					{
+					from_normal_to_fec(conn_info,0,0,out_n,out_arr,out_len,out_delay);
+					}
+
+					conn_info.stat.report_as_server(ip_port);
+					continue;
+				}
+				else
+				{
+					assert(conn_info.conv_manager.is_u64_used(fd64));
+
+					conv=conn_info.conv_manager.find_conv_by_u64(fd64);
+					conn_info.conv_manager.update_active_time(conv);
+					conn_info.update_active_time();
+
+					int fd=fd_manager.to_fd(fd64);
+					data_len=recv(fd,data,max_data_len,0);
+
+					mylog(log_trace,"received a packet from udp_fd,len:%d,conv=%d\n",data_len,conv);
+
+					if(data_len<0)
+					{
+						mylog(log_debug,"udp fd,recv_len<0 continue,%s\n",strerror(errno));
+
+						continue;
+					}
+
+					if(!disable_mtu_warn&&data_len>=mtu_warn)
+					{
+						mylog(log_warn,"huge packet,data len=%d (>=%d).strongly suggested to set a smaller mtu at upper level,to get rid of this warn\n ",data_len,mtu_warn);
+					}
+
+					char * new_data;
+					int new_len;
+					put_conv(conv,data,data_len,new_data,new_len);
+
+					from_normal_to_fec(conn_info,new_data,new_len,out_n,out_arr,out_len,out_delay);
+				}
+
+				mylog(log_trace,"out_n=%d\n",out_n);
+				for(int i=0;i<out_n;i++)
+				{
+					delay_send(out_delay[i],dest,out_arr[i],out_len[i]);
+				}
+				//mylog(log_trace,"[%s] send packet\n",ip_port.to_s());
+
+			}
+			else
+			{
+				mylog(log_fatal,"unknown fd,this should never happen\n");
+				myexit(-1);
+			}
+		}
+		delay_manager.check();
+	}
+	return 0;
+}
--- a/tunnel.h
+++ b/tunnel.h
@@ -0,0 +1,17 @@
+/*
+ * tunnel.h
+ *
+ *  Created on: Oct 26, 2017
+ *      Author: root
+ */
+
+#ifndef TUNNEL_H_
+#define TUNNEL_H_
+
+
+#include "misc.h"
+
+int tunnel_client_event_loop();
+int tunnel_server_event_loop();
+
+#endif /* TUNNEL_H_ */
Author	SHA1	Message	Date
wangyu-	6f55b8a2fc	fix help page	2017-11-19 06:11:06 -06:00
wangyu-	85122b7155	help page	2017-11-19 06:07:40 -06:00
wangyu-	0cef4503be	changed default mode to mode 0;improve help page	2017-11-19 06:02:20 -06:00
wangyu-	744c8f56ae	Update README.zh-cn.md	2017-11-19 02:02:44 -08:00
wangyu-	8680b06882	Update README.zh-cn.md	2017-11-18 05:31:01 -08:00
wangyu-	57cce3a424	Update README.zh-cn.md	2017-11-18 05:30:23 -08:00
wangyu-	6ab4294bd9	Update README.zh-cn.md	2017-11-18 05:22:27 -08:00
wangyu-	ebdabb4598	Update README.zh-cn.md	2017-11-18 05:21:41 -08:00
wangyu-	c254545290	Update README.zh-cn.md	2017-11-18 01:14:33 -08:00
wangyu-	1301fdc96d	Merge branch 'master' of https://github.com/wangyu-/UDPspeeder	2017-11-18 00:05:25 -06:00
wangyu-	f58dfe4fdd	fixed a warn of last commit	2017-11-18 00:03:08 -06:00
wangyu-	3fd198a1e2	fixed a possible MTU issue when redundant num=0	2017-11-18 00:01:21 -06:00
wangyu-	017e851c2d	Update README.md	2017-11-17 09:50:58 -08:00
wangyu-	2929eead0e	Update README.md	2017-11-17 09:46:52 -08:00
wangyu-	e2c2f2dfec	Update README.zh-cn.md	2017-11-17 09:12:30 -08:00
wangyu-	d0ebd67dbd	Update README.md	2017-11-17 08:54:30 -08:00
wangyu-	211fcee8db	Update README.zh-cn.md	2017-11-17 08:51:05 -08:00
wangyu-	6b4a2dfe21	increased timeout	2017-11-17 01:17:31 -06:00
wangyu-	65939eb9e7	Merge branch 'master' of https://github.com/wangyu-/UDPspeeder	2017-11-17 01:12:17 -06:00
wangyu-	9c00a2f3a6	increased timeout	2017-11-17 01:12:04 -06:00
wangyu-	4706e51fad	Update README.zh-cn.md	2017-11-16 10:52:01 -08:00
wangyu-	c3786a91d8	Update README.zh-cn.md	2017-11-16 10:31:11 -08:00
wangyu-	680f127350	Update README.zh-cn.md	2017-11-16 10:28:57 -08:00
wangyu-	7713cf0c15	Update README.zh-cn.md	2017-11-16 10:18:16 -08:00
wangyu-	5024a02d20	Update README.zh-cn.md	2017-11-16 10:12:32 -08:00
wangyu-	f2d9deb2a3	Update README.zh-cn.md	2017-11-16 10:09:47 -08:00
wangyu-	c5c440245c	Update README.zh-cn.md	2017-11-16 06:42:20 -08:00
wangyu-	c95e2db1af	Update README.zh-cn.md	2017-11-16 06:41:26 -08:00
wangyu-	82a2b3cb9a	Update README.zh-cn.md	2017-11-16 03:10:37 -08:00
wangyu-	8ecc696da5	Update README.zh-cn.md	2017-11-16 03:04:21 -08:00
wangyu-	6e6fc62fb8	Update README.zh-cn.md	2017-11-15 20:34:32 -08:00
wangyu-	ea03e5f8ec	Update README.zh-cn.md	2017-11-15 20:33:21 -08:00
wangyu-	3be2642397	Update README.zh-cn.md	2017-11-15 20:31:47 -08:00
wangyu-	d05bcfc2f8	Update README.zh-cn.md	2017-11-15 20:28:48 -08:00
wangyu-	e50292bd4a	Update README.zh-cn.md	2017-11-15 20:26:42 -08:00
wangyu-	363aec12d4	Update README.zh-cn.md	2017-11-12 20:40:01 -08:00
wangyu-	4364a7b789	Update README.zh-cn.md	2017-11-12 20:38:37 -08:00
wangyu-	17a42a1218	Update README.zh-cn.md	2017-11-11 14:33:20 -08:00
wangyu-	9152fef474	Update README.zh-cn.md	2017-11-11 14:29:46 -08:00
wangyu-	eb7406e6f3	Update README.zh-cn.md	2017-11-11 14:18:42 -08:00
wangyu-	296a84be3c	Update README.zh-cn.md	2017-11-11 14:12:58 -08:00
wangyu-	69558ab16c	Update README.zh-cn.md	2017-11-11 00:58:25 -08:00
wangyu-	37b7eb7282	Update README.zh-cn.md	2017-11-09 01:47:08 -08:00
wangyu-	4ca57cf656	Update README.zh-cn.md	2017-11-09 01:11:19 -08:00
wangyu-	1a744d69ea	Update README.zh-cn.md	2017-11-08 15:59:24 -08:00
wangyu-	86b68697e1	Update README.zh-cn.md	2017-11-08 15:57:23 -08:00
wangyu-	63a2f923ab	Update README.md	2017-11-08 15:54:54 -08:00
wangyu-	5774b478fc	Update README.md	2017-11-08 15:54:12 -08:00
wangyu-	e91dcf6647	Update README.zh-cn.md	2017-11-08 15:50:31 -08:00
wangyu-	f6c97af854	Update README.md	2017-11-08 15:49:59 -08:00
wangyu-	2f097d720d	changed a log	2017-11-08 03:59:36 -06:00
wangyu-	c6de39acc8	relax an assert	2017-11-08 03:53:13 -06:00
wangyu-	0f425a3a98	added version.txt	2017-11-08 03:38:23 -06:00
wangyu-	2cc560088b	Update README.md	2017-11-07 02:58:34 -08:00
wangyu-	b83b21f2eb	Merge branch 'master' of https://github.com/wangyu-/UDPspeeder	2017-11-07 03:55:01 -06:00
wangyu-	f526046846	do not drop packet in mode 1	2017-11-07 03:54:54 -06:00
wangyu-	1be5fb5fc1	Update README.zh-cn.md	2017-11-06 18:53:00 -08:00
wangyu-	27874aa413	changed parameter	2017-11-06 03:19:28 -06:00
wangyu-	fde2329528	Merge branch 'master' of https://github.com/wangyu-/UDPspeeder	2017-11-06 03:17:38 -06:00
wangyu-	30b78c1339	changed parameter	2017-11-06 03:17:34 -06:00
wangyu-	17a3074b34	Update README.zh-cn.md	2017-11-05 03:38:20 -08:00
wangyu-	cce7d18163	Update README.zh-cn.md	2017-11-04 21:50:37 -07:00
wangyu-	90029e1ea7	Update README.zh-cn.md	2017-11-03 22:50:24 -07:00
wangyu-	c26a4708ee	Update README.zh-cn.md	2017-11-03 22:49:28 -07:00
wangyu-	fdbf4ab205	Update udpspeeder_openvpn.md	2017-11-03 22:46:33 -07:00
wangyu-	7ee72ee84c	Update udpspeeder_openvpn.md	2017-11-03 22:45:39 -07:00
wangyu-	cac5f8271c	Update README.zh-cn.md	2017-10-31 21:19:26 -07:00
wangyu-	219e8e2592	Update README.zh-cn.md	2017-10-31 21:18:33 -07:00
wangyu-	02b4d59500	Update README.zh-cn.md	2017-10-31 21:18:02 -07:00
wangyu-	894bf7cd0d	Update README.zh-cn.md	2017-10-31 20:52:50 -07:00
wangyu-	a8cab43139	Update README.zh-cn.md	2017-10-31 20:52:31 -07:00
wangyu-	369a9d8e1b	Update README.zh-cn.md	2017-10-31 20:51:39 -07:00
wangyu-	0f7894f42f	revert last comit	2017-10-31 22:29:16 -05:00
wangyu-	d7daae25bf	relax restriction for fec_data_num==1	2017-10-31 22:14:57 -05:00
wangyu-	d7271b74b4	Merge branch 'master' of https://github.com/wangyu-/UDPspeeder	2017-10-31 22:05:58 -05:00
wangyu-	1eeef14470	update log	2017-10-31 22:05:51 -05:00
wangyu-	9d30caffc9	Add files via upload	2017-11-01 10:35:01 +08:00
wangyu-	e50e383ede	Delete ping_compare_cn.PNG	2017-10-31 19:34:47 -07:00
wangyu-	c73a197971	Update README.zh-cn.md	2017-10-31 19:31:24 -07:00
wangyu-	ba44617124	Add files via upload	2017-11-01 10:30:10 +08:00
wangyu-	19e718f0a3	new function tcp csum	2017-10-31 19:43:25 -05:00
wangyu-	bd323be9c4	fixed log	2017-10-30 20:29:04 -05:00
wangyu-	35a85e38ae	fixed a missed continue	2017-10-30 11:23:37 -05:00
wangyu-	4871983b8f	Update README.zh-cn.md	2017-10-30 08:40:17 -07:00
wangyu-	fec3c39cad	Update README.zh-cn.md	2017-10-30 08:38:18 -07:00
wangyu-	4d55348a4d	Update README.md	2017-10-30 08:35:41 -07:00
wangyu-	36eef02d8a	Update README.md	2017-10-30 08:35:24 -07:00
wangyu-	3fd48dd20e	Update README.md	2017-10-30 08:33:56 -07:00
wangyu-	766aa274d9	Update README.md	2017-10-30 08:33:36 -07:00
wangyu-	d961537682	Update udpspeeder_openvpn.md	2017-10-29 23:42:21 -07:00
wangyu-	4dcfe87bd3	Update README.zh-cn.md	2017-10-29 20:47:23 -07:00
wangyu-	2aa0ece465	Update README.md	2017-10-29 20:46:24 -07:00
wangyu-	05fd84b675	Update README.zh-cn.md	2017-10-29 03:41:28 -07:00
wangyu-	1e00ff7c87	add option keep-reconnect	2017-10-28 08:56:24 -05:00
wangyu-	c76818c6a8	trival	2017-10-28 07:44:25 -05:00
wangyu-	e3e2fe6d36	changed default sub-net	2017-10-28 06:48:24 -05:00
wangyu-	7cd87e7b59	trival	2017-10-28 04:36:37 -05:00
wangyu-	c3aae9579b	refactor	2017-10-28 02:05:14 -05:00
wangyu-	d9f80781a2	changed --tun-dev to required argument	2017-10-28 01:45:50 -05:00
wangyu-	203604e1b5	Merge branch 'master' of https://github.com/wangyu-/UDPspeeder	2017-10-28 01:03:12 -05:00
wangyu-	5ac6e30c8e	removed tun_dev.cpp and .h	2017-10-28 01:02:58 -05:00
wangyu-	6728d5a561	add dest ip restriction	2017-10-28 00:45:19 -05:00
wangyu-	f9eaae172f	fixed typo	2017-10-27 23:19:08 -05:00
wangyu-	a0eeedb12a	more log,more robust	2017-10-27 22:51:33 -05:00
wangyu-	324d34c87a	Update README.zh-cn.md	2017-10-27 20:25:17 -07:00
wangyu-	1fe64b5689	Update README.zh-cn.md	2017-10-27 20:20:35 -07:00
wangyu-	d5e900e91c	aloww len==fec_mtu	2017-10-27 21:16:59 -05:00
wangyu-	8e00273db4	just commit	2017-10-27 08:28:43 -05:00
wangyu-	73d68a2baf	just commit	2017-10-27 05:23:46 -05:00
wangyu-	d1c88bbc07	implemented connection reject	2017-10-27 04:58:20 -05:00
wangyu-	caa8320743	implement dev tun	2017-10-27 00:03:35 -05:00
wangyu-	f64108867c	fixed merge conflict	2017-10-26 09:49:46 -05:00
wangyu-	716e5458f9	Merge pull request #48 from bg6cq/master typo fix	2017-10-26 22:34:48 +08:00
Zhang Huanjie	91c9a9e705	usage typo fix	2017-10-26 22:09:05 +08:00
Zhang Huanjie	e633b5b3ce	typo fix	2017-10-26 21:59:47 +08:00
wangyu-	47d762958f	refactor	2017-10-26 08:54:18 -05:00
wangyu-	0a34cb47ab	refactor	2017-10-26 08:38:52 -05:00
wangyu-	cdba2f8c2c	Update README.zh-cn.md	2017-10-26 01:23:24 -07:00
wangyu-	8f9deafce7	Update README.zh-cn.md	2017-10-26 01:21:59 -07:00
wangyu-	d6e445f220	Merge pull request #47 from Carmot/master Minor memory issues fixed.	2017-10-26 14:43:00 +08:00
Jose Angel Gariburo	3d2ec1bdf2	fec.c is external so leaving it as it is. Changed log output to 'log_warn'.	2017-10-26 08:29:04 +02:00
wangyu-	c0450160d6	trival	2017-10-26 00:31:21 -05:00
wangyu-	b23dd44aa7	Update README.md	2017-10-25 21:04:25 -07:00
wangyu-	fc540c6fe2	Update README.md	2017-10-25 21:01:08 -07:00
wangyu-	c0b57e66fd	Update README.zh-cn.md	2017-10-25 11:35:15 -07:00
wangyu-	00f8ddf48e	Update README.zh-cn.md	2017-10-25 11:22:04 -07:00
wangyu-	63cc9a3c5d	Update README.zh-cn.md	2017-10-25 11:21:44 -07:00
wangyu-	47c5823f52	Update README.zh-cn.md	2017-10-25 11:16:22 -07:00
wangyu-	5351095f42	Update README.zh-cn.md	2017-10-25 11:15:45 -07:00
wangyu-	64b0420ff4	Update README.zh-cn.md	2017-10-25 10:33:08 -07:00
wangyu-	695b0e46a8	Update README.zh-cn.md	2017-10-25 10:24:08 -07:00
wangyu-	1798529d28	Update README.zh-cn.md	2017-10-25 10:22:51 -07:00
wangyu-	f50e0d4dc6	Update README.zh-cn.md	2017-10-25 09:57:18 -07:00
wangyu-	c0817fba47	Update README.zh-cn.md	2017-10-25 09:43:06 -07:00
wangyu-	0a6e9f500a	Update README.zh-cn.md	2017-10-25 09:41:58 -07:00
Jose Angel Gariburo	c8fb35cf53	Minor memory issues fixed. 'u_long' variable name changed to 'u_dword'. It is not a good idea to name a variable as an existing type.	2017-10-25 17:48:22 +02:00
wangyu-	9d62ea410f	Update README.zh-cn.md	2017-10-25 00:58:11 -07:00
wangyu-	d1c36cc197	Update README.zh-cn.md	2017-10-25 00:55:12 -07:00
wangyu-	a99791c98a	Update README.zh-cn.md	2017-10-25 00:33:45 -07:00
wangyu-	0e1b86b237	Merge branch 'master' of https://github.com/wangyu-/UDPspeeder	2017-10-25 02:29:43 -05:00
wangyu-	a1cddaf671	change log	2017-10-25 02:29:11 -05:00