netlink

Netlink

内核和用户层通信的方式
在内核态主动发起消息，而用户态的程序可以阻塞等待消息。

• 自己独立的地址编码

    struct sockaddr_nl  
    {  
        sa_family_t nl_family;          //AF_NETLINK, PF_NETLINK  
        unsigned short  nl_pad;             //必须为0  
        __u32       nl_pid;             //通信端口,一般使用用户进程PID  
        								//当是内核的地址，该值必须用0， sendto时，这个对端地址为0	
    	__u32       nl_groups;           //组播掩码  
    };

• 创建socket的时候，需要指定 netlink的通信协议号，支持自定义

socket(PF_NETLINK, SOCK_DGRAM | SOCK_CLOEXEC, NETLINK_KOBJECT_UEVENT);

    #define NETLINK_ROUTE       0   /* Routing/device hook              */  
    #define NETLINK_UNUSED      1   /* Unused number                */  
    #define NETLINK_USERSOCK    2   /* Reserved for user mode socket protocols  */  
    #define NETLINK_FIREWALL    3   /* Firewalling hook             */  
    #define NETLINK_INET_DIAG   4   /* INET socket monitoring           */  
    #define NETLINK_NFLOG       5   /* netfilter/iptables ULOG */  
    #define NETLINK_XFRM        6   /* ipsec */  
    #define NETLINK_SELINUX     7   /* SELinux event notifications */  
    #define NETLINK_ISCSI       8   /* Open-iSCSI */  
    #define NETLINK_AUDIT       9   /* auditing */  
    #define NETLINK_FIB_LOOKUP  10    
    #define NETLINK_CONNECTOR   11  
    #define NETLINK_NETFILTER   12  /* netfilter subsystem */  
    #define NETLINK_IP6_FW      13  
    #define NETLINK_DNRTMSG     14  /* DECnet routing messages */  
    #define NETLINK_KOBJECT_UEVENT  15  /*！！！！ Kernel messages to userspace */  
    #define NETLINK_GENERIC     16  
    /* leave room for NETLINK_DM (DM Events) */  
    #define NETLINK_SCSITRANSPORT   18  /* SCSI Transports */  
    #define NETLINK_ECRYPTFS    19  

• recvfrom sendto
  每个消息都附带一个netlink消息头，struct nlmsghdr；

struct nlmsghdr {
_u32 nlmsg_len; /* Length of msg including header */
_u32 nlmsg_type; /* 操作命令 */
_u16 nlmsg_flags; /* various flags */
_u32 nlmsg_seq; /* Sequence number */
_u32 nlmsg_pid; /* 进程PID */
};

   0                   1                   2                   3
   0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
   |                          Length                             |
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
   |            Type              |           Flags              |
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
   |                      Sequence Number                        |
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
   |                      Process ID (PID)                       |
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

• recvmsg sendmsg
  1.标记
  2.套接口地址与长度
  3.分散写入
  4.附属数据

返回值 msg_iov数据总的大小

ssize_t sendto(int sockfd, const void *buf, size_t len, int flags,
              const struct sockaddr *dest_addr, socklen_t addrlen);

ssize_t sendmsg(int sockfd, const struct msghdr *msg, int flags);

msg_name 和 msg_namelen 表示地址相当于dest_addr，addrlen
msg_iov，msg_iovlen表示传输的数据相当于 buf，len
msg_control 辅助数据
msg_flags只用于recvmsg
MSG_EOR 当接收到记录结尾时会设置这一位。
MSG_TRUNC 这个标记位表明数据的结尾被截短，因为接收缓冲区太小不足以接收全部的数据。
MSG_CTRUNC 这个标记位表明某些控制数据(附属数据)被截短，因为缓冲区太小。
MSG_OOB 这个标记位表明接收了带外数据。
MSG_ERRQUEUE 这个标记位表明没有接收到数据，但是返回一个扩展错误。

struct msghdr {
    void          *msg_name;        // protocol address
    socklen_t      msg_namelen;     // size of protocol address
    struct iovec  *msg_iov;         // scatter/gather array
    int            msg_iovlen;      // elements in msg_iov
    void          *msg_control;     // ancillary data (cmsghdr struct)
    socklen_t      msg_controllen;  // length of ancillary data
    int            msg_flags;       // flags returned by recvmsg()
};

//在一次函数调用中读、写多个非连续缓冲区
//散布读（scatter read）和聚集写（gather write）
struct iovec {
    ptr_t iov_base; /* Starting address */
    size_t iov_len; /* Length in bytes */
};
//iovcnt 传输的iovec的数量
ssize_t readv(int filedes, const struct iovec *iov, int iovcnt);
ssize_t writev(int filedes, const struct iovec *iov, int iovcnt);

附属数据
bionic/libc/include/sys/socket.h

struct cmsghdr { 
    socklen_t cmsg_len; 
    int       cmsg_level; 
    int       cmsg_type; 
};

//四个字节对齐
#define CMSG_ALIGN(len) ( ((len)+sizeof(long)-1) & ~(sizeof(long)-1) )

#define CMSG_LEN(len) (CMSG_ALIGN(sizeof(struct cmsghdr)) + (len))
cmsghdr头结构+len

#define CMSG_SPACE(len) (CMSG_ALIGN(sizeof(struct cmsghdr)) + CMSG_ALIGN(len)
总的空间cmsghdr头结构长度 + len  --4字节对齐

#define CMSG_DATA(cmsg) ((void*)((char*)(cmsg) + CMSG_ALIGN(sizeof(struct cmsghdr))))
指针指向跟随在头部以及填充字节之后的附属数据的第一个字节

#define CMSG_FIRSTHDR(msg) \
  ((msg)->msg_controllen >= sizeof(struct cmsghdr) \
   ? (struct cmsghdr*) (msg)->msg_control : (struct cmsghdr*) NULL)
 附属数据缓冲区内的第一个附属对象的struct cmsghdr
{//例子

}
#define CMSG_NXTHDR(mhdr, cmsg) __cmsg_nxthdr((mhdr), (cmsg))
 下一个struct cmsghdr

对齐的原理
(len + 0b0000,0000,0000,0000,0000,0000,0000,0011) & 0b1111,1111,1111,1111,1111,1111,1111,1100

   #define alignment_down(a, size) (a & (~(size-1)) )
   #define alignment_up(a, size)   ((a+size-1) & (~ (size-1)))
   #define _ALIGN(addr,size) (((addr)+(size)-1)&(~((size)-1)))
   #define PAGE_SIZE         4096
   #define PAGE_MASK         (~(PAGE_SIZE-1))
   #define PAGE_ALIGN(addr) -(((addr)+PAGE_SIZE-1) & PAGE_MASK)

uevent中的例子

 char control[CMSG_SPACE(sizeof(struct ucred))];   //control申请这么多空间
 struct msghdr hdr = {
        &addr,
        sizeof(addr),
        &iov,
        1,
        control,
        sizeof(control),
        0,
    };
     ssize_t n = recvmsg(socket, &hdr, 0);
    struct cmsghdr *cmsg = CMSG_FIRSTHDR(&hdr);
struct ucred *cred = (struct ucred *)CMSG_DATA(cmsg);