RSTP（Rapid Spanning Tree Protocol，快速生成树协议）是STP（Spanning Tree Protocol，生成树协议）的一个改进版本，主要区别在于RSTP通过引入新的机制来加快网络的收敛速度。具体来说：

• 端口角色的增加和优化：RSTP增加了Backup端口和边缘端口两种角色。共有5中端口角色，而STP只有3种。这些额外的端口角色简化了生成树协议的理解和部署。
• 端口状态的优化：RSTP将端口状态从STP的5种减少到3种，分别是:Discarding、Learning和Forwarding状态，这种简化使得端口状态的装换更加快速和直接。
• BPDE（Bridge Protocol Data Unit）格式和处理的优化：RSTP改变了BPDU的格式，充分利用了BPDU中Flag字段，明确了端口角色，并增加了P/A机制位的标识，这种优化提高了BPDU的处理效率，从而加快了网络的收敛速度。
• 收敛速度的优化：RSTP引入了边缘端口机制、根端口快速切换机制以及Proposal/Agreement机制，这些机制共同作用，显著提高了网络在拓扑变化时的收敛速度。相比之下，STP的收敛依赖于计时器，且收敛速度较慢。

STP协议端口状态

STP端口的物种状态分别是：

Disabled（禁用）：端口处于禁用状态，不参与STP计算，也不转发数据帧。

Blocking（阻塞）：端口接收数据帧，但不允许转发这些数据帧，只用于接收BPDU（Bridge Protocol Data Unit，桥协议数据单元）、执行STP计算。

Listening（监听）：端口处理监听状态，接收数据帧，并且准备进行状态转移。

Learning（学习）：端口开始从其他接口学习到的数据中学习MAC地址，并添加到端口的MAC地址表中。

Forwarding（转发）：端口转发数据帧，作为桥接网络中的一个活动端口，可以转发数据帧到其他端口。

STP环路检测原理

设备通过发送环路检测报文并检测其是否返回本设备（不要求收、发端口为同一端口）以确认是否存在环路，若某端口收到了由本设备发出的环路检测报文，就认定该端口所在链路存在环路。

BPDU的主要参数

Root Identifier（根ID）：发送此配置BPDU的交换机所认为的根交换机的交换机标识。

Root Path Cost(到根的路径开销)：从发送此配置BPDU的交换机到达根交换机的最短路径总开销，含交换机根端口的开销，不含发送此配置的BPDU的端口的开销。

Bridge Identifier（桥ID）：发送此配置BPDU的交换机的STP交换机标识。

Port Identifier（端口ID）：发送此配置的BPDU的交换机端口的STP端口标识

比较次序：Root Identifier > Root Path Cost> Bridge Identifier>Port Identifier

值越小越优先

Bridge ID

• 桥ID是交换机的STP标示符，一共8个字节，由2个自己的优先级和6个字节的MAC地址构成；
• 桥优先级缺省为32768，可以手工修改。
• MAC地址为交换机的背板MAC
• 网络中Bridge ID最小的交换价将成为根桥；

Path cost

路径开销是一个端口量，是STP/RSTP协议用于选择链路的参考值；

端口路径开销的默认值及取值范围由选定的路径开销算法决定，路径开销与端口的带宽成反比。

华为网络设备支持的路径开销计算标准：802.1d-1998、802.1t、legacy等。

Port ID

•端口ID （2字节）= 端口优先级（1字节）+ 端口编号（1字节）

•缺省优先级128，范围0-255，越小越优。

选举过程

下面以一个实例进行说明

从上图中可以看到，运行了生成树算法之后，S4选择阻塞F0/0，S5选择阻塞F0/2，应该是看到非根桥转发的BPDU后做的选择

非根桥转发的BPDU

传统生成树中只有根桥才能产生配置BPDU，非根桥只是转发根桥产生的BPDU。试想一下如果仅仅只是转发根桥的BPDU，不对BPDU做任何更改，就像转发常规数据帧一样。这样的话S4和S5怎么判断阻塞的端口，又凭啥要阻塞我的端口。这就引出了关键点---非根桥转发的BPDU。

可以看出非根桥转发根桥BPDU时做了如下修改：

1，根路径开销；

2，发送者网桥ID；

3，端口标识；

通过三个步骤来分解S4和S5阻塞端口的过程

第一步：S2和S3看到的BPDU

S2和S3看到报文后，发现自己的MAC地址不如人家，就放弃竞选根桥了，只好去选根端口了，只要接收不到更小的BID，那就赶紧定下来根端口，报文中的根路径开销决定了S2和S3的根端口。

第二步：S4看到的BPDU

当S4看到BPDU后发现自己和根桥的ID比，根本不如人家，那我只好来定夺根端口了，S2告诉我到S1的开销是19，S3告诉我到S1的开销也是19，那当然是选择S2的那条路了（S2的端口ID小于S3的端口ID），S3的这条路只能堵上。网桥ID决定了S4的根端口，路径开销决定了阻塞端口。

第三步：S5看到的BPDU

当S5看到途中的BPDU后也没什么想法了，MAC地址比根桥的大太多了，只好选根端口了。S5的F0/3和F0/2收到S4发来的BPDU报文里唯一的不同只有端口ID，那就只有比较端口ID了，哪个端口接收到的端口ID大就阻塞该端口。端口ID决定了S5的根端口，路径开销决定了阻塞端口。

如何判断内核支持STP还是RSTP?

从内核中定义的宏可以看出，内核中只支持STP，并不支持RSTP，RSTP需要在用户层来实现：

STP协议头：

RSTP协议头

MSTP

多生成树协议MSTP（Multiple Spanning Tree Protocol）是IEEE 802.1s中定义的一种新型生成树协议。该协议主要是解决了传统stp和rstp，在同一局域网内所有的vlan共享一个生成树，无法在vlan间实现数据流量的负载均衡，以及产生次优路径的问题。

在每个广播域中选举出一个根网桥

在选举开始的时候，所有的交换机都会认为自己是根桥，他们产生并相互发送自己的BPDU。比较BPDU重Bridge-id:优先级+MAC地址，越小越优先。

Bridge-id: 优先级+MAC地址 优先级：范围0-65535，必须设置成 4096的倍数，最小是0 优先级的默认值是 32768, 越小越优先.

当根桥选举完成后，只有根桥自己能够产生BPDU，其他的非根桥负责接收和转发来自根桥的BPDU。

在非根交换机上，选出根端口

比较到达根桥的路径开销

如下图，SW1的端口2比端口1路径开销小，SW2的端口2比端口1开销小。开销小的端口为根端口。

比较Sender-BID

如果开销值相同：比较端口接收到的BPDU中 Sender-BID这一参数。例如此图，SW4的两个接口到根桥的开销值是相同的，比较Sender-BID，端口2的对端交换机BPDU更优先，从而SW4的端口2是根端口

 比较Sender-PID

如果交换机的各个接口接收到BPDU中的 Sender-BID也相同，则比较 BPDU中的 Sender-PID，例如此图，SW5的两个端口，对端交换机的3端口PID更优先，所以SW5的2端口选为根端口。

1. 如果交换机的各个接口接收到BPDU中的 Sender-PID也相同，例如使用了HUB集线器，则比较交换机本身端口的PID

比较交换机本身端口的PID

如果交换机的各个接口接收到BPDU中的 Sender-PID也相同，例如使用了HUB集线器，则比较交换机本身端口的PID

选出指定端口

在剩余的链路上，选举出指定端口

1）根端口对面的端口，就是指定端口；

2）在剩余链路上，选举指定端口

选举方法如下：

• 比较端口所在交换机到达根桥的路径开销，越小越优先
• 如果端口所在交换机到根桥的开销相同，则比较端口所在交换机的BID
• 如果所在交换机的BID也相同，则比较交换机的PID

应用层实现

STP

执行brctl show命令可以看到STP协议是否开启。

root@jenet:~# brctl show
bridge name     bridge id               STP enabled     interfaces
br0             8000.5c857ea0aab9       no              lan
                                                        wan

STP协议是在内核中实现的，brctl通过API获取内核参数：

static int br_cmd_show(int argc, char *const* argv)
{
	int i;

	printf("bridge name\tbridge id\t\tSTP enabled\tinterfaces\n");
	if (argc == 1)
		br_foreach_bridge(show_bridge, NULL);
	else
		for(i = 2; i <= argc; i++)
			show_bridge(argv[i - 1], NULL);

	return 0;
}

在如下函数中获取桥信息时，一并获取STP启用状态：

info->stp_enabled = fetch_int(path, "stp_state");

int br_get_bridge_info(const char *bridge, struct bridge_info *info)
{
	DIR *dir;
	char path[SYSFS_PATH_MAX];

	snprintf(path, SYSFS_PATH_MAX, SYSFS_CLASS_NET "%s/bridge", bridge);
	dir = opendir(path);
	if (dir == NULL) {
		dprintf("path '%s' is not a directory\n", path);
		goto fallback;
	}

	memset(info, 0, sizeof(*info));
	fetch_id(path, "root_id", &info->designated_root);
	fetch_id(path, "bridge_id", &info->bridge_id);
	info->root_path_cost = fetch_int(path, "root_path_cost");
	fetch_tv(path, "max_age", &info->max_age);
	fetch_tv(path, "hello_time", &info->hello_time);
	fetch_tv(path, "forward_delay", &info->forward_delay);
	fetch_tv(path, "max_age", &info->bridge_max_age);
	fetch_tv(path, "hello_time", &info->bridge_hello_time);
	fetch_tv(path, "forward_delay", &info->bridge_forward_delay);
	fetch_tv(path, "ageing_time", &info->ageing_time);
	fetch_tv(path, "hello_timer", &info->hello_timer_value);
	fetch_tv(path, "tcn_timer", &info->tcn_timer_value);
	fetch_tv(path, "topology_change_timer", 
		 &info->topology_change_timer_value);;
	fetch_tv(path, "gc_timer", &info->gc_timer_value);

	info->root_port = fetch_int(path, "root_port");
	info->stp_enabled = fetch_int(path, "stp_state");
	info->topology_change = fetch_int(path, "topology_change");
	info->topology_change_detected = fetch_int(path, "topology_change_detected");

	closedir(dir);
	return 0;

fallback:
	return old_get_bridge_info(bridge, info);
}

在设备中查看如下：

RSTP

RSTP启动涉及到三个文件：

rstpctl：rstp客户端，用来和rstpd通信。

rstpd: rstp协议的守护进程，用来处理BPDT。

bridge-stp：通过rstpctl来管理rstpd的脚本文件。

rstpctl

#define RSTP_SERVER_SOCK_NAME ".rstp_server"

在ctl_client_init（）函数中创建一个unix套接字连接rstpd进程的socket。

int ctl_client_init(void)
{
	struct sockaddr_un sa_svr;
	int s;
	TST(strlen(RSTP_SERVER_SOCK_NAME) < sizeof(sa_svr.sun_path), -1);

	s = socket(PF_UNIX, SOCK_DGRAM, 0);
	if (s < 0) {
		ERROR("Couldn't open unix socket: %m");
		return -1;
	}

	set_socket_address(&sa_svr, RSTP_SERVER_SOCK_NAME);

	struct sockaddr_un sa;
	char tmpname[64];
	sprintf(tmpname, "RSTPCTL_%d", getpid());
	set_socket_address(&sa, tmpname);
	/* We need this bind. The autobind on connect isn't working properly.
	   The server doesn't get a proper sockaddr in recvmsg if we don't do this.
	 */
	if (bind(s, (struct sockaddr *)&sa, sizeof(sa)) != 0) {
		ERROR("Couldn't bind socket: %m");
		close(s);
		return -1;
	}

	if (connect(s, (struct sockaddr *)&sa_svr, sizeof(sa_svr)) != 0) {
		ERROR("Couldn't connect to server");
		close(s);
		return -1;
	}
	fd = s;

	return 0;
}

 通过rstpctl设置命令的处理函数定义如下：

static const struct command commands[] = {
	{0, 32, "showbridge", cmd_showbridge,
	 "[<bridge> ... ]\t\tshow bridge state"},
	{1, 32, "showport", cmd_showport,
	 "<bridge> [<port> ... ]\tshow port state"},
	{1, 32, "showportdetail", cmd_showportdetail,
	 "<bridge> [<port> ... ]\tshow port state (detail)"},
	{2, 0, "rstp", cmd_rstp,
	 "<bridge> {on|off}\tenable/disable rstpd control"},
	{2, 0, "setbridgestate", cmd_setbridgestate,
	 "<bridge> {on|off}\tstart/stop rstp (when enabled)"},
	{2, 0, "setbridgeprio", cmd_setbridgeprio,
	 "<bridge> <priority>\tset bridge priority (0-61440)"},
	{2, 0, "sethello", cmd_setbridgehello,
	 "<bridge> <hellotime>\tset bridge hello time (1-10)"},
	{2, 0, "setmaxage", cmd_setbridgemaxage,
	 "<bridge> <maxage>\tset bridge max age (6-40)"},
	{2, 0, "setfdelay", cmd_setbridgefdelay,
	 "<bridge> <fwd_delay>\tset bridge forward delay (4-30)"},
	{2, 0, "setforcevers", cmd_setbridgeforcevers,
	 "<bridge> {normal|slow}\tnormal RSTP or force to STP"},
	{3, 0, "setportprio", cmd_setportprio,
	 "<bridge> <port> <priority>\tset port priority (0-240)"},
	{3, 0, "setportpathcost", cmd_setportpathcost,
	 "<bridge> <port> <cost>\tset port path cost"},
	{3, 0, "setportedge", cmd_setportedge,
	 "<bridge> <port> {yes|no}\tconfigure if it is an edge port"},
	{3, 0, "setportnonstp", cmd_setportnonstp,
	 "<bridge> <port> {yes|no}\tdisable STP for the port"},
	{3, 0, "setportp2p", cmd_setportp2p,
	 "<bridge> <port> {yes|no|auto}\tset whether p2p connection"},
	{2, 0, "portmcheck", cmd_portmcheck,
	 "<bridge> <port>\ttry to get back from STP to RSTP mode"},
	{1, 0, "debuglevel", cmd_debuglevel, "<level>\t\tLevel of verbosity"},
};

开启rstp命令的函数为：

static int cmd_rstp(int argc, char *const *argv)
{
	int stp, r;
	int br_index = get_index(argv[1], "bridge");

	if (!strcmp(argv[2], "on") || !strcmp(argv[2], "yes")
	    || !strcmp(argv[2], "1"))
		stp = 1;
	else if (!strcmp(argv[2], "off") || !strcmp(argv[2], "no")
		 || !strcmp(argv[2], "0"))
		stp = 0;
	else {
		fprintf(stderr, "expect on/off for argument\n");
		return 1;
	}
	r = CTL_enable_bridge_rstp(br_index, stp);
	if (r) {
		fprintf(stderr, "Failed to enable/disable RSTP: %s\n",
			CTL_error_explanation(r));
		return -1;
	}
	return 0;
}

如果开启了STP，则开启RSTP会失败

在init_bridge_stp函数中判断是否开启了STP

int init_bridge_stp(struct ifdata *br)
{
	if (br->stp_up) {
		ERROR("STP already started");
		return 0;
	}

	/* Init STP state */
	TST(init_rstplib_instance(br) == 0, -1);

	struct ifdata *p = br->port_list;
	while (p) {
		if (add_port_stp(p) != 0)
			break;
		p = p->port_next;
	}
	if (p) {
		struct ifdata *q = br->port_list;
		while (q != p) {
			remove_port_stp(q);
			q = q->port_next;
		}
		/* Clear bridge STP state */
		clear_rstplib_instance(br);
		return -1;
	}
	br->stp_up = 1;
	return 0;
}

debuglevel

#define LOG_LEVEL_NONE 0

#define LOG_LEVEL_ERROR 1

#define LOG_LEVEL_INFO  2

#define LOG_LEVEL_DEBUG 3

#define LOG_LEVEL_RSTPLIB 4

bridge-stp脚本

#!/bin/bash
#
# Script to start/stop spanning tree called from kernel
# Make sure umask is sane
umask 022

# Set up a default search path.
PATH="/sbin:/usr/sbin:/bin:/usr/bin"
export PATH

if [ $# -ne 2 ]; then
   echo "Usage: bridge-stp <bridge> {start|stop}"
   exit 1
fi
bridge=$1
service=rstpd
pid_file=/var/run/${service}.pid

# Set $pid to pids from /var/run* for {program}.  $pid should be declared
# local in the caller.
# Returns LSB exit code for the 'status' action.
checkpid() {
        pid=
        if [ -f "$1" ] ; then
                local line p
                read line < "$pid_file"
                for p in $line ; do
                        [ -z "${p//[0-9]/}" -a -d "/proc/$p" ] && pid="$pid $p"
                done
                if [ -n "$pid" ]; then
                        return 0
                fi
                return 1 # "Program is dead and /var/run pid file exists"
        fi
        return 3 # "Program is not running"
}

daemon() {
    local pid
    checkpid $pid_file

    [ -n "$pid" ] && return
    # start it
    /sbin/rstpd
    RETVAL=$?
    [ $RETVAL -eq 0 ] && touch /var/lock/subsys/$service
}

start() {
    daemon
}

case $2 in
     start)
        daemon
        exec /sbin/rstpctl rstp $bridge on ;;
     stop)
        exec /sbin/rstpctl rstp $bridge off ;;
     *)
        echo "Unknown action:" $2
        echo "Usage: bridge-stp <bridge> {start|stop}"
        exit 1

该脚本先检查pid文件，如果不存在则启动/sbin/rstpd。但是在daemon函数中，如果pid为空，则直接返回了，

RSTPD

rstpd通过rtnl和内核通信

内核中对STP报文的处理

在kernel/net/bridge/br.c文件中注册了bridge模块:

module_init(br_init)
module_exit(br_deinit)
MODULE_LICENSE("GPL");
MODULE_VERSION(BR_VERSION);
MODULE_ALIAS_RTNL_LINK("bridge");

TCN BPDU

Topology Change Notification BPDU

STP与RSTP的冲突问题

在使用RSTP前，需要先通过 brctl stp br0 off命令关闭内核中的STP功能。此时打印内核的br->stp_enabled的值为0，执行brctl stp br0 on命令后，打印内核的br->stp_enabled的值为1。

先执行brctl stp br0 off，再执行bridge-stp br0 start开启rstpd功能，此时打印内核的br->stp_enabled的值为0，也就是说内核的br->stp_enabled只代表STP的开关，和RSTP无关。

先执行brctl stp br0 on命令开启内核的STP功能，再执行bridge-stp br0 start开启RSTP功能，在PC上抓包，发现设备发了两份包，一份是STP，一份是RSTP：

执行brctl stp br0 on报错

[ 1881.348918] br-lan: failed to start userspace STP (-8)
[ 2126.204139] br-lan: failed to start userspace STP (-8)

在执行该命令时会调用如下函数：

函数调用关系：

net/bridge/br_ioctl.c

old_dev_ioctl()-->br_stp_set_enabled（）-->br_stp_start()-->

查看错误码-8，是说文件格式有错：

发现是由于在bridge-stp文件中增加了打印:set -xe导致的，将其去掉后，报256的错误：

此错误是由于在/var/run/目录下已经存在rstpd.pid文件，删除此文件后，再次执行brctl stp br0 on命令，发现终端进程卡住了，证明是在拉起用户态进程的处理过程中异常了。

如果bridge-stp文件格式错误，则打印如下信息：

root@jenet:~# brctl stp br0 on
[   65.292720] br0: br_stp_call_user: pro:/sbin/bridge-stp,br->dev->name:root@jenet:~# br0, arg:start, rc=-8
[   65.292745] br_stp_start: err=-8
[   65.292751] br0: failed to start userspace STP (-8)
[   65.292756] br_stp_start: using kernel STP
^C

转发时状态判断

在linux-4.14.195/net/bridge/br_forward.c文件中有如下判断：

只有在端口状态为forwarding时才会转发数据，在此处增加打印：

端口状态的定义如下：

#define BR_STATE_DISABLED 0

#define BR_STATE_LISTENING 1

#define BR_STATE_LEARNING 2

#define BR_STATE_FORWARDING 3

#define BR_STATE_BLOCKING 4

当产生环路时，后台有如下日志打印：

[  122.990840] net_ratelimit: 2302 callbacks suppressed

该条打印在linux-4.14.195/net/core/utils.c文件中定义：

/*
 * All net warning printk()s should be guarded by this function.
 */
int net_ratelimit(void)
{
	return __ratelimit(&net_ratelimit_state);
}
EXPORT_SYMBOL(net_ratelimit);

#define __ratelimit(state) ___ratelimit(state, __func__)

最终在/linux-4.14.195/lib/ratelimit.c 中定义如下：

/*
 * __ratelimit - rate limiting
 * @rs: ratelimit_state data
 * @func: name of calling function
 *
 * This enforces a rate limit: not more than @rs->burst callbacks
 * in every @rs->interval
 *
 * RETURNS:
 * 0 means callbacks will be suppressed.
 * 1 means go ahead and do it.
 */
int ___ratelimit(struct ratelimit_state *rs, const char *func)
{
	unsigned long flags;
	int ret;

	if (!rs->interval)
		return 1;

	/*
	 * If we contend on this state's lock then almost
	 * by definition we are too busy to print a message,
	 * in addition to the one that will be printed by
	 * the entity that is holding the lock already:
	 */
	if (!raw_spin_trylock_irqsave(&rs->lock, flags))
		return 0;

	if (!rs->begin)
		rs->begin = jiffies;

	if (time_is_before_jiffies(rs->begin + rs->interval)) {
		if (rs->missed) {
			if (!(rs->flags & RATELIMIT_MSG_ON_RELEASE)) {
				printk_deferred(KERN_WARNING
						"%s: %d callbacks suppressed\n",
						func, rs->missed);
				rs->missed = 0;
			}
		}
		rs->begin   = jiffies;
		rs->printed = 0;
	}
	if (rs->burst && rs->burst > rs->printed) {
		rs->printed++;
		ret = 1;
	} else {
		rs->missed++;
		ret = 0;
	}
	raw_spin_unlock_irqrestore(&rs->lock, flags);

	return ret;
}
EXPORT_SYMBOL(___ratelimit);

#define RATELIMIT_STATE_INIT(name, interval_init, burst_init) {		\
		.lock		= __RAW_SPIN_LOCK_UNLOCKED(name.lock),	\
		.interval	= interval_init,			\
		.burst		= burst_init,				\
	}

interval被初始化为：5 * HZ

burst被初始化为：10

系统每2秒收到一个BPDU报文：

[  178.952397] File: net/bridge/br_stp.c, Line: 515
[  180.952265] File: net/bridge/br_stp.c, Line: 515
[  182.952208] File: net/bridge/br_stp.c, Line: 515
[  184.952296] File: net/bridge/br_stp.c, Line: 515
[  186.952215] File: net/bridge/br_stp.c, Line: 515
[  188.952118] File: net/bridge/br_stp.c, Line: 515
[  190.952155] File: net/bridge/br_stp.c, Line: 515
[  192.952016] File: net/bridge/br_stp.c, Line: 515
[  194.951912] File: net/bridge/br_stp.c, Line: 515
[  196.951908] File: net/bridge/br_stp.c, Line: 515
[  198.951927] File: net/bridge/br_stp.c, Line: 515
[  200.951827] File: net/bridge/br_stp.c, Line: 515
[  202.951748] File: net/bridge/br_stp.c, Line: 515
[  204.951707] File: net/bridge/br_stp.c, Line: 515
[  206.951666] File: net/bridge/br_stp.c, Line: 515
[  208.951704] File: net/bridge/br_stp.c, Line: 515

当端口进入blocking状态后，会进入下面的分支：

admin@OpenWrt:/# [  364.949717] File: net/bridge/br_stp.c, Line: 515
[  366.949706] File: net/bridge/br_stp.c, Line: 515
[  368.949802] File: net/bridge/br_stp.c, Line: 515
[  370.949674] File: net/bridge/br_stp.c, Line: 515
[  372.315630] mtk_soc_eth 1e100000.ethernet eth0: port 2 link up
[  372.949774] File: net/bridge/br_stp.c, Line: 515
[  372.958992] File: net/bridge/br_stp.c, Line: 473, br:br-lan, dev;lan
[  372.971657] br-lan: topology change detected, sending tcn bpdu
[  372.983329] File: net/bridge/br_stp.c, Line: 413, br:br-lan, dev;lan
[  372.995995] File: net/bridge/br_stp.c, Line: 44, br:br-lan, dev;lan
[  373.008480] File: toolchain ./include/net/switchdev.h, Line: 202, dev;lan
[  373.022002] File: net/bridge/br_stp.c, Line: 54, br:br-lan, dev;lan
[  373.034487] br-lan: port 2(lan) entered blocking state
[  373.044732] File: net/bridge/br_netlink.c, Line: 468, dev:lan
[  373.056272] File: net/bridge/br_stp.c, Line: 515
[  373.065487] File: net/bridge/br_stp.c, Line: 473, br:br-lan, dev;lan
[  373.909721] File: net/bridge/br_stp.c, Line: 515
[  373.918930] File: net/bridge/br_stp.c, Line: 473, br:br-lan, dev;lan
[  373.931590] File: toolchain ./include/net/switchdev.h, Line: 202, dev;br-lan
[  374.949629] File: net/bridge/br_stp.c, Line: 515
[  374.958833] File: net/bridge/br_stp.c, Line: 473, br:br-lan, dev;lan
[  374.971519] File: net/bridge/br_stp.c, Line: 515
[  374.980721] File: net/bridge/br_stp.c, Line: 473, br:br-lan, dev;lan
[  376.949633] File: net/bridge/br_stp.c, Line: 515
[  376.958838] File: net/bridge/br_stp.c, Line: 473, br:br-lan, dev;lan
[  376.971522] File: net/bridge/br_stp.c, Line: 515
[  376.980726] File: net/bridge/br_stp.c, Line: 473, br:br-lan, dev;lan
[  377.665397] p->state != BR_STATE_FORWARDING File: net/bridge/br_forward.c, Line: 173, br:br-lan, dev:lan, state:4
[  377.685838] p->state != BR_STATE_FORWARDING File: net/bridge/br_forward.c, Line: 31, br:br-lan, dev:lan, state:4
[  378.668114] p->state != BR_STATE_FORWARDING File: net/bridge/br_forward.c, Line: 173, br:br-lan, dev:lan, state:4
[  378.688567] p->state != BR_STATE_FORWARDING File: net/bridge/br_forward.c, Line: 31, br:br-lan, dev:lan, state:4
[  378.949712] File: net/bridge/br_stp.c, Line: 515
[  378.958916] File: net/bridge/br_stp.c, Line: 473, br:br-lan, dev;lan
[  378.971590] File: net/bridge/br_stp.c, Line: 515
[  378.980773] File: net/bridge/br_stp.c, Line: 473, br:br-lan, dev;lan
[  379.674625] p->state != BR_STATE_FORWARDING File: net/bridge/br_forward.c, Line: 173, br:br-lan, dev:lan, state:4
[  379.695070] p->state != BR_STATE_FORWARDING File: net/bridge/br_forward.c, Line: 31, br:br-lan, dev:lan, state:4
[  380.684415] p->state != BR_STATE_FORWARDING File: net/bridge/br_forward.c, Line: 173, br:br-lan, dev:lan, state:4
[  380.704867] p->state != BR_STATE_FORWARDING File: net/bridge/br_forward.c, Line: 31, br:br-lan, dev:lan, state:4

对应的代码为：

此时通过命令行查看接口状态为blocking:

调用maybe_deliver函数的地方为：

修改pathcost

默认情况下，每个端口的pathcost都是100，此时端口ID大的端口被blocking :

此时，修改wan口的pathcost为1000，再查看时发现wan口被blocking了。

每次收到BPDU报文都会更新根端口和指定端口，在br_port_state_selection（）函数中，只要不是根端口和指定端口，都会进入到blocking状态：

br_forward.c文件中函数调用关系为：br_flood()-->maybe_deliver()-->should_deliver()

在上面的代码中增加打印：

打印如下：

[  139.790465] br_port_state_selection: designated_port dev:wan,designated_port:32770,port_id:32770
[  139.790507] br_port_state_selection: designated_port dev:lan,designated_port:32769,port_id:32769
[  139.790556] br0: port 1(lan) entered listening state
[  141.628637] br_port_state_selection: designated_port dev:wan,designated_port:32770,port_id:32770
[  141.628700] br_port_state_selection: root_port dev:lan,port_no:1,root_port:1
[  142.992958] br_port_state_selection: designated_port dev:wan,designated_port:32770,port_id:32770
[  142.993017] br_port_state_selection: root_port dev:lan,port_no:1,root_port:1
[  144.089529] br_port_state_selection: designated_port dev:wan,designated_port:32770,port_id:32770
[  144.089558] br_port_state_selection: root_port dev:lan,port_no:1,root_port:1
^C
root@jenet:~# [  146.006292] br_port_state_selection: designated_port dev:wan,designated_port:32770,port_id:32770
[  146.006321] br_port_state_selection: root_port dev:lan,port_no:1,root_port:1
[  148.139787] br_port_state_selection: designated_port dev:wan,designated_port:32770,port_id:32770
[  148.139822] br_port_state_selection: root_port dev:lan,port_no:1,root_port:1
[  150.059831] br_port_state_selection: designated_port dev:wan,designated_port:32770,port_id:32770
[  150.059929] br_port_state_selection: root_port dev:lan,port_no:1,root_port:1
[  152.196543] br_port_state_selection: designated_port dev:wan,designated_port:32770,port_id:32770
[  152.196623] br_port_state_selection: root_port dev:lan,port_no:1,root_port:1
[  154.116565] br_port_state_selection: designated_port dev:wan,designated_port:32770,port_id:32770
[  154.116607] br_port_state_selection: root_port dev:lan,port_no:1,root_port:1
[  155.071072] br0: port 1(lan) entered learning state
[  156.033369] br_port_state_selection: designated_port dev:wan,designated_port:32770,port_id:32770
[  156.033429] br_port_state_selection: root_port dev:lan,port_no:1,root_port:1
[  158.028491] rk_gmac-dwmac fe010000.ethernet lan: Link is Down
[  158.028826] br0: port 1(lan) entered disabled state
[  158.029008] br_port_state_selection: designated_port dev:wan,designated_port:32770,port_id:32770
[  158.029058] br0: topology change detected, propagating

先插入wan口网线，再插入lan口网线

使能STP日志

lan口不插网线，wan口插网线，日志如下：

root@jenet:~# brctl  stp br0 on
[   65.363143] br0: br_stp_call_user: pro:/sbin/bridge-stp,br->dev->name:brroot@jenet:~# 0, arg:start, rc=-8
[   65.363174] br_stp_start: err=-8
[   65.363182] br0: failed to start userspace STP (-8)
[   65.363189] br_stp_start: using kernel STP,br->stp_enabled=1
[   65.363199] br_port_state_selection: dev:wan,designated_port:32770,port_id:32770,port_no:2,root_port:0

组网测试

有线口环路

两台RK的LAN与LAN对接，WAN与WAN对接。在一台

另外一台的端口状态为：

先插入WAN口，打印如下：

先插入wan口，再插入lan口：

root@jenet:~# [  151.997410] rk_gmac-dwmac fe2a0000.ethernet wan: Link is Up - 1Gbps/Full - flow control off
[  151.997612] br0: port 2(wan) entered blocking state
[  151.997652] br_port_state_selection: dev:wan,designated_port:32770,port_id:32770,port_no:2,root_port:0
[  151.997684] p->designated_bridge: dev:wan 8000.5c857ea0aab9
[  151.997713] p->br->bridge_id: dev:wan 8000.5c857ea0aab9
[  151.997740] br0: port 2(wan) entered listening state
[  152.553989] br_port_state_selection: dev:wan,designated_port:32770,port_id:32770,port_no:2,root_port:2
[  152.554018] p->designated_bridge: dev:wan 8000.5c857e031866
[  152.554024] p->br->bridge_id: dev:wan 8000.5c857ea0aab9
[  153.702511] br_port_state_selection: dev:wan,designated_port:32770,port_id:32770,port_no:2,root_port:2
[  153.702542] p->designated_bridge: dev:wan 8000.5c857e031866
[  153.702550] p->br->bridge_id: dev:wan 8000.5c857ea0aab9
[  155.622649] br_port_state_selection: dev:wan,designated_port:32770,port_id:32770,port_no:2,root_port:2
[  155.622683] p->designated_bridge: dev:wan 8000.5c857e031866
[  155.622690] p->br->bridge_id: dev:wan 8000.5c857ea0aab9

root@jenet:~# [  157.755927] br_port_state_selection: dev:wan,designated_port:32770,port_id:32770,port_no:2,root_port:2
[  157.755952] p->designated_bridge: dev:wan 8000.5c857e031866
[  157.755960] p->br->bridge_id: dev:wan 8000.5c857ea0aab9
[  159.676130] br_port_state_selection: dev:wan,designated_port:32770,port_id:32770,port_no:2,root_port:2
[  159.676184] p->designated_bridge: dev:wan 8000.5c857e031866
[  159.676195] p->br->bridge_id: dev:wan 8000.5c857ea0aab9
[  161.596176] br_port_state_selection: dev:wan,designated_port:32770,port_id:32770,port_no:2,root_port:2
[  161.596232] p->designated_bridge: dev:wan 8000.5c857e031866
[  161.596252] p->br->bridge_id: dev:wan 8000.5c857ea0aab9
[  163.729525] br_port_state_selection: dev:wan,designated_port:32770,port_id:32770,port_no:2,root_port:2
[  163.729573] p->designated_bridge: dev:wan 8000.5c857e031866
[  163.729588] p->br->bridge_id: dev:wan 8000.5c857ea0aab9
[  165.649643] br_port_state_selection: dev:wan,designated_port:32770,port_id:32770,port_no:2,root_port:2
[  165.649687] p->designated_bridge: dev:wan 8000.5c857e031866
[  165.649695] p->br->bridge_id: dev:wan 8000.5c857ea0aab9
[  166.074219] rk_gmac-dwmac fe010000.ethernet lan: Link is Up - 1Gbps/Full - flow control off
[  166.074292] IPv6: ADDRCONF(NETDEV_CHANGE): lan: link becomes ready
[  166.074453] br0: port 1(lan) entered blocking state
[  166.074479] br_port_state_selection: dev:wan,designated_port:32770,port_id:32770,port_no:2,root_port:2
[  166.074491] p->designated_bridge: dev:wan 8000.5c857e031866
[  166.074502] p->br->bridge_id: dev:wan 8000.5c857ea0aab9
[  166.074514] br_port_state_selection: dev:lan,designated_port:32769,port_id:32769,port_no:1,root_port:2
[  166.074525] p->designated_bridge: dev:lan 8000.5c857ea0aab9
[  166.074536] p->br->bridge_id: dev:lan 8000.5c857ea0aab9
[  166.074544] br0: port 1(lan) entered listening state
[  167.059879] br0: port 2(wan) entered learning state
[  167.573021] br_port_state_selection: dev:wan,designated_port:32770,port_id:32770,port_no:2,root_port:2
[  167.573079] p->designated_bridge: dev:wan 8000.5c857e031866
[  167.573100] p->br->bridge_id: dev:wan 8000.5c857ea0aab9
[  167.573128] br_port_state_selection: dev:lan,designated_port:32769,port_id:32769,port_no:1,root_port:2
[  167.573218] p->designated_bridge: dev:lan 8000.5c857ea0aab9
[  167.573253] p->br->bridge_id: dev:lan 8000.5c857ea0aab9
[  167.573472] br_port_state_selection: dev:wan,designated_port:32770,port_id:32770,port_no:2,root_port:1
[  167.573511] p->designated_bridge: dev:wan 8000.5c857e031866
[  167.573543] p->br->bridge_id: dev:wan 8000.5c857ea0aab9
[  167.573569] br0: topology change detected, sending tcn bpdu
[  167.573625] br0: port 2(wan) entered blocking state
[  167.573749] br_port_state_selection: dev:lan,designated_port:32769,port_id:32769,port_no:1,root_port:1
[  167.573789] p->designated_bridge: dev:lan 8000.5c857e031866
[  167.573821] p->br->bridge_id: dev:lan 8000.5c857ea0aab9
[  168.556317] br_port_state_selection: dev:wan,designated_port:32770,port_id:32770,port_no:2,root_port:1
[  168.556347] p->designated_bridge: dev:wan 8000.5c857e031866
[  168.556353] p->br->bridge_id: dev:wan 8000.5c857ea0aab9
[  168.556359] br_port_state_selection: dev:lan,designated_port:32769,port_id:32769,port_no:1,root_port:1
[  168.556365] p->designated_bridge: dev:lan 8000.5c857e031866
[  168.556371] p->br->bridge_id: dev:lan 8000.5c857ea0aab9
[  169.703044] br_port_state_selection: dev:wan,designated_port:32770,port_id:32770,port_no:2,root_port:1
[  169.703078] p->designated_bridge: dev:wan 8000.5c857e031866
[  169.703085] p->br->bridge_id: dev:wan 8000.5c857ea0aab9
[  169.703091] br_port_state_selection: dev:lan,designated_port:32769,port_id:32769,port_no:1,root_port:1
[  169.703097] p->designated_bridge: dev:lan 8000.5c857e031866
[  169.703103] p->br->bridge_id: dev:lan 8000.5c857ea0aab9
[  169.703274] br_port_state_selection: dev:wan,designated_port:32770,port_id:32770,port_no:2,root_port:1
[  169.703291] p->designated_bridge: dev:wan 8000.5c857e031866
[  169.703298] p->br->bridge_id: dev:wan 8000.5c857ea0aab9
[  169.703304] br_port_state_selection: dev:lan,designated_port:32769,port_id:32769,port_no:1,root_port:1
[  169.703310] p->designated_bridge: dev:lan 8000.5c857e031866
[  169.703316] p->br->bridge_id: dev:lan 8000.5c857ea0aab9
[  171.623480] br_port_state_selection: dev:wan,designated_port:32770,port_id:32770,port_no:2,root_port:1
[  171.623559] p->designated_bridge: dev:wan 8000.5c857e031866
[  171.623589] p->br->bridge_id: dev:wan 8000.5c857ea0aab9
[  171.623618] br_port_state_selection: dev:lan,designated_port:32769,port_id:32769,port_no:1,root_port:1
[  171.623670] p->designated_bridge: dev:lan 8000.5c857e031866
[  171.623712] p->br->bridge_id: dev:lan 8000.5c857ea0aab9
[  171.623832] br_port_state_selection: dev:wan,designated_port:32770,port_id:32770,port_no:2,root_port:1
[  171.623873] p->designated_bridge: dev:wan 8000.5c857e031866
[  171.623923] p->br->bridge_id: dev:wan 8000.5c857ea0aab9
[  171.623965] br_port_state_selection: dev:lan,designated_port:32769,port_id:32769,port_no:1,root_port:1
[  171.623992] p->designated_bridge: dev:lan 8000.5c857e031866
[  171.624035] p->br->bridge_id: dev:lan 8000.5c857ea0aab9
[  173.756637] br_port_state_selection: dev:wan,designated_port:32770,port_id:32770,port_no:2,root_port:1
[  173.756693] p->designated_bridge: dev:wan 8000.5c857e031866
[  173.756713] p->br->bridge_id: dev:wan 8000.5c857ea0aab9
[  173.756731] br_port_state_selection: dev:lan,designated_port:32769,port_id:32769,port_no:1,root_port:1
[  173.756751] p->designated_bridge: dev:lan 8000.5c857e031866
[  173.756768] p->br->bridge_id: dev:lan 8000.5c857ea0aab9
[  173.756849] br_port_state_selection: dev:wan,designated_port:32770,port_id:32770,port_no:2,root_port:1
[  173.756885] p->designated_bridge: dev:wan 8000.5c857e031866
[  173.756915] p->br->bridge_id: dev:wan 8000.5c857ea0aab9
[  173.756943] br_port_state_selection: dev:lan,designated_port:32769,port_id:32769,port_no:1,root_port:1
[  173.756970] p->designated_bridge: dev:lan 8000.5c857e031866
[  173.756996] p->br->bridge_id: dev:lan 8000.5c857ea0aab9
[  175.676768] br_port_state_selection: dev:wan,designated_port:32770,port_id:32770,port_no:2,root_port:1
[  175.676813] p->designated_bridge: dev:wan 8000.5c857e031866
[  175.676821] p->br->bridge_id: dev:wan 8000.5c857ea0aab9
[  175.676827] br_port_state_selection: dev:lan,designated_port:32769,port_id:32769,port_no:1,root_port:1
[  175.676833] p->designated_bridge: dev:lan 8000.5c857e031866
[  175.676851] p->br->bridge_id: dev:lan 8000.5c857ea0aab9
[  175.676903] br_port_state_selection: dev:wan,designated_port:32770,port_id:32770,port_no:2,root_port:1
[  175.676916] p->designated_bridge: dev:wan 8000.5c857e031866
[  175.676931] p->br->bridge_id: dev:wan 8000.5c857ea0aab9
[  175.676942] br_port_state_selection: dev:lan,designated_port:32769,port_id:32769,port_no:1,root_port:1
[  175.676952] p->designated_bridge: dev:lan 8000.5c857e031866
[  175.676961] p->br->bridge_id: dev:lan 8000.5c857ea0aab9
[  177.596891] br_port_state_selection: dev:wan,designated_port:32770,port_id:32770,port_no:2,root_port:1
[  177.596950] p->designated_bridge: dev:wan 8000.5c857e031866
[  177.596971] p->br->bridge_id: dev:wan 8000.5c857ea0aab9
[  177.596989] br_port_state_selection: dev:lan,designated_port:32769,port_id:32769,port_no:1,root_port:1

p->designated_bridge：存储的是网络中协商出来、指定的桥ID

p->br->bridge_id：此端口本地的桥ID。

 只有跟桥设备会发送BPfDU报文，所以根桥上不会频繁打印上述日志，根桥设备分别插入wan/lan口网线打印日志为：

root@jenet:~# brctl show
bridge name     bridge id               STP enabled     interfaces
br0             8000.5c857e031866       yes             lan
                                                        wan
root@jenet:~# [  500.544494] rk_gmac-dwmac fe010000.ethernet lan: Link is Up - 1Gbps/Full - flow control off
[  500.544611] IPv6: ADDRCONF(NETDEV_CHANGE): lan: link becomes ready
[  500.544828] br0: port 1(lan) entered blocking state
[  500.544852] br_port_state_selection: dev:wan,designated_port:32770,port_id:32770,port_no:2,root_port:0
[  500.544866] p->designated_bridge: dev:wan 8000.5c857e031866
[  500.544877] p->br->bridge_id: dev:wan 8000.5c857e031866
[  500.544895] br_port_state_selection: dev:lan,designated_port:32769,port_id:32769,port_no:1,root_port:0
[  500.544906] p->designated_bridge: dev:lan 8000.5c857e031866
[  500.544913] p->br->bridge_id: dev:lan 8000.5c857e031866
[  500.544924] br0: port 1(lan) entered listening state
[  501.530741] br0: port 1(lan) received tcn bpdu
[  501.530786] br0: topology change detected, propagating

root@jenet:~#
root@jenet:~#
root@jenet:~#
root@jenet:~#
root@jenet:~# syspara show

所以说根桥设备上的所有端口都不会阻塞。

二层VPN环路测试

环路探测OK，tap口处于blocking状态：

现场拓扑测试

RK1表项

RK1使用WAN口上网（172.16.67.164），表项为：

接口MAC为：

root@jenet:~# ifconfig
br0: flags=4163<UP,BROADCAST,RUNNING,MULTICAST>  mtu 1500
        inet 192.168.1.10  netmask 255.255.255.0  broadcast 192.168.1.255
        inet6 fe80::4a43  prefixlen 128  scopeid 0x20<link>
        inet6 fe80::5e85:7eff:fea0:aab9  prefixlen 64  scopeid 0x20<link>
        ether 5c:85:7e:a0:aa:b9  txqueuelen 1000  (Ethernet)
        RX packets 1514  bytes 303055 (295.9 KiB)
        RX errors 0  dropped 0  overruns 0  frame 0
        TX packets 3401  bytes 204716 (199.9 KiB)
        TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0

br0:0: flags=4163<UP,BROADCAST,RUNNING,MULTICAST>  mtu 1500
        inet 10.99.0.9  netmask 255.255.0.0  broadcast 10.99.255.255
        ether 5c:85:7e:a0:aa:b9  txqueuelen 1000  (Ethernet)

eth1: flags=4099<UP,BROADCAST,MULTICAST>  mtu 1500
        ether ee:32:17:2d:d8:c3  txqueuelen 1000  (Ethernet)
        RX packets 2  bytes 636 (636.0 B)
        RX errors 0  dropped 0  overruns 0  frame 0
        TX packets 7  bytes 1130 (1.1 KiB)
        TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0

lan: flags=4163<UP,BROADCAST,RUNNING,MULTICAST>  mtu 1500
        inet6 fe80::5e85:7eff:fea0:aab9  prefixlen 64  scopeid 0x20<link>
        ether 5c:85:7e:a0:aa:b9  txqueuelen 1000  (Ethernet)
        RX packets 2097  bytes 303934 (296.8 KiB)
        RX errors 0  dropped 0  overruns 0  frame 0
        TX packets 2528  bytes 123534 (120.6 KiB)
        TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0
        device interrupt 39

lo: flags=73<UP,LOOPBACK,RUNNING>  mtu 65536
        inet 127.0.0.1  netmask 255.0.0.0
        inet6 ::1  prefixlen 128  scopeid 0x10<host>
        loop  txqueuelen 1000  (Local Loopback)
        RX packets 5596  bytes 474120 (463.0 KiB)
        RX errors 0  dropped 0  overruns 0  frame 0
        TX packets 5596  bytes 474120 (463.0 KiB)
        TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0

tap0: flags=4163<UP,BROADCAST,RUNNING,MULTICAST>  mtu 1500
        inet6 fe80::9888:2cff:feef:9a88  prefixlen 64  scopeid 0x20<link>
        ether 9a:88:2c:ef:9a:88  txqueuelen 100  (Ethernet)
        RX packets 3218  bytes 356161 (347.8 KiB)
        RX errors 0  dropped 0  overruns 0  frame 0
        TX packets 1720  bytes 78406 (76.5 KiB)
        TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0

wan: flags=4163<UP,BROADCAST,RUNNING,MULTICAST>  mtu 1500
        inet 172.16.67.164  netmask 255.255.255.0  broadcast 172.16.67.255
        ether 5c:85:7e:a0:aa:ba  txqueuelen 1000  (Ethernet)
        RX packets 17714  bytes 1909382 (1.8 MiB)
        RX errors 0  dropped 0  overruns 0  frame 0
        TX packets 5665  bytes 3572406 (3.4 MiB)
        TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0
        device interrupt 52

RK2表项

RK2使用5G上网，表项为：

root@jenet:~# ifconfig
br0: flags=4163<UP,BROADCAST,RUNNING,MULTICAST>  mtu 1500
        inet 192.168.0.1  netmask 255.255.255.0  broadcast 192.168.0.255
        inet6 fe80::5e85:7eff:fe03:1866  prefixlen 64  scopeid 0x20<link>
        inet6 fe80::4a43  prefixlen 128  scopeid 0x20<link>
        ether 5c:85:7e:03:18:66  txqueuelen 1000  (Ethernet)
        RX packets 3246  bytes 459630 (448.8 KiB)
        RX errors 0  dropped 0  overruns 0  frame 0
        TX packets 28  bytes 2707 (2.6 KiB)
        TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0

br0:0: flags=4163<UP,BROADCAST,RUNNING,MULTICAST>  mtu 1500
        inet 10.99.0.2  netmask 255.255.0.0  broadcast 10.99.255.255
        ether 5c:85:7e:03:18:66  txqueuelen 1000  (Ethernet)

eth1: flags=4163<UP,BROADCAST,RUNNING,MULTICAST>  mtu 1500
        inet 10.12.94.121  netmask 255.255.255.0  broadcast 10.12.94.255
        inet6 2408:8441:810:2b25:43f:aff:fe0a:21c8  prefixlen 64  scopeid 0x0<global>
        ether 06:3f:0a:0a:21:c8  txqueuelen 1000  (Ethernet)
        RX packets 2780  bytes 1757658 (1.6 MiB)
        RX errors 0  dropped 0  overruns 0  frame 0
        TX packets 1985  bytes 264222 (258.0 KiB)
        TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0

lan: flags=4163<UP,BROADCAST,RUNNING,MULTICAST>  mtu 1500
        inet6 fe80::5e85:7eff:fe03:1866  prefixlen 64  scopeid 0x20<link>
        ether 5c:85:7e:03:18:66  txqueuelen 1000  (Ethernet)
        RX packets 2467  bytes 161356 (157.5 KiB)
        RX errors 0  dropped 0  overruns 0  frame 0
        TX packets 2693  bytes 482909 (471.5 KiB)
        TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0
        device interrupt 39

lo: flags=73<UP,LOOPBACK,RUNNING>  mtu 65536
        inet 127.0.0.1  netmask 255.0.0.0
        inet6 ::1  prefixlen 128  scopeid 0x10<host>
        loop  txqueuelen 1000  (Local Loopback)
        RX packets 1264  bytes 152711 (149.1 KiB)
        RX errors 0  dropped 0  overruns 0  frame 0
        TX packets 1264  bytes 152711 (149.1 KiB)
        TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0

tap0: flags=4163<UP,BROADCAST,RUNNING,MULTICAST>  mtu 1500
        inet6 fe80::f0d0:31ff:fea5:7a82  prefixlen 64  scopeid 0x20<link>
        ether f2:d0:31:a5:7a:82  txqueuelen 100  (Ethernet)
        RX packets 2668  bytes 481922 (470.6 KiB)
        RX errors 0  dropped 0  overruns 0  frame 0
        TX packets 1904  bytes 128361 (125.3 KiB)
        TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0

wan: flags=4099<UP,BROADCAST,MULTICAST>  mtu 1500
        ether 5c:85:7e:03:18:67  txqueuelen 1000  (Ethernet)
        RX packets 0  bytes 0 (0.0 B)
        RX errors 0  dropped 0  overruns 0  frame 0
        TX packets 0  bytes 0 (0.0 B)
        TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0
        device interrupt 52

root@jenet:~#

RK3表项

RK3使用WAN口上网（172.16.67.162），表项为：

接口MAC为：

root@jenet:~# ifconfig
br0: flags=4163<UP,BROADCAST,RUNNING,MULTICAST>  mtu 1500
        inet 192.168.0.1  netmask 255.255.255.0  broadcast 192.168.0.255
        inet6 fe80::4a43  prefixlen 128  scopeid 0x20<link>
        inet6 fe80::188d:5fff:feb5:8d09  prefixlen 64  scopeid 0x20<link>
        ether 1a:8d:5f:b5:8d:09  txqueuelen 1000  (Ethernet)
        RX packets 1706  bytes 293831 (286.9 KiB)
        RX errors 0  dropped 0  overruns 0  frame 0
        TX packets 563  bytes 48032 (46.9 KiB)
        TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0

br0:0: flags=4163<UP,BROADCAST,RUNNING,MULTICAST>  mtu 1500
        inet 10.99.0.5  netmask 255.255.0.0  broadcast 10.99.255.255
        ether 1a:8d:5f:b5:8d:09  txqueuelen 1000  (Ethernet)

lan: flags=4163<UP,BROADCAST,RUNNING,MULTICAST>  mtu 1500
        inet6 fe80::188d:5fff:feb5:8d09  prefixlen 64  scopeid 0x20<link>
        ether 1a:8d:5f:b5:8d:09  txqueuelen 1000  (Ethernet)
        RX packets 1773  bytes 329602 (321.8 KiB)
        RX errors 0  dropped 4  overruns 0  frame 0
        TX packets 1857  bytes 130455 (127.3 KiB)
        TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0
        device interrupt 39

lo: flags=73<UP,LOOPBACK,RUNNING>  mtu 65536
        inet 127.0.0.1  netmask 255.0.0.0
        inet6 ::1  prefixlen 128  scopeid 0x10<host>
        loop  txqueuelen 1000  (Local Loopback)
        RX packets 834  bytes 50500 (49.3 KiB)
        RX errors 0  dropped 0  overruns 0  frame 0
        TX packets 834  bytes 50500 (49.3 KiB)
        TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0

tap0: flags=4163<UP,BROADCAST,RUNNING,MULTICAST>  mtu 1500
        inet6 fe80::3418:1eff:fec4:84ea  prefixlen 64  scopeid 0x20<link>
        ether 36:18:1e:c4:84:ea  txqueuelen 100  (Ethernet)
        RX packets 950  bytes 64213 (62.7 KiB)
        RX errors 0  dropped 0  overruns 0  frame 0
        TX packets 1442  bytes 274346 (267.9 KiB)
        TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0

wan: flags=4163<UP,BROADCAST,RUNNING,MULTICAST>  mtu 1500
        inet 172.16.67.162  netmask 255.255.255.0  broadcast 172.16.67.255
        ether 1a:8d:5f:b5:8d:09  txqueuelen 1000  (Ethernet)
        RX packets 5380  bytes 503835 (492.0 KiB)
        RX errors 0  dropped 3  overruns 0  frame 0
        TX packets 2714  bytes 3365557 (3.2 MiB)
        TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0
        device interrupt 52

将RK2和RK3连接

RK2的tap0口被block住，环路探测有效。

 STP的hairpin模式

root@jenet:~# brctl --help
Usage: brctl [commands]
commands:
        addbr           <bridge>                add bridge
        delbr           <bridge>                delete bridge
        addif           <bridge> <device>       add interface to bridge
        delif           <bridge> <device>       delete interface from bridge
        hairpin         <bridge> <port> {on|off}        turn hairpin on/off
        setageing       <bridge> <time>         set ageing time

int br_set_hairpin_mode(const char *bridge, const char *port, int hairpin_mode)
{
	return port_set(bridge, port, "hairpin_mode", hairpin_mode, 0);
}

最终使写在了如下地方：

搜索代码，并没有地方判断此标志位：