ebpf的sockmap调用strp_init来设置回调函数
static int smap_init_sock(struct smap_psock *psock,
struct sock *sk)
{
static const struct strp_callbacks cb = {
.rcv_msg = smap_read_sock_strparser,
.parse_msg = smap_parse_func_strparser,
.read_sock_done = smap_read_sock_done,
};
return strp_init(&psock->strp, sk, &cb);
}
static void smap_read_sock_strparser(struct strparser *strp,
struct sk_buff *skb)
{
struct smap_psock *psock;
rcu_read_lock();
psock = container_of(strp, struct smap_psock, strp);
smap_do_verdict(psock, skb);
rcu_read_unlock();
}
strp_init初始化
strp_init会设置回调函数同时设置一个定时器，一条消息超时没有接收完整后将被调用。
int strp_init(struct strparser *strp, struct sock *csk,
struct strp_callbacks *cb)
{
struct socket *sock = csk->sk_socket;
if (!cb || !cb->rcv_msg || !cb->parse_msg)
return -EINVAL;
if (!sock->ops->read_sock || !sock->ops->peek_len)
return -EAFNOSUPPORT;
memset(strp, 0, sizeof(*strp));
strp->sk = csk;
setup_timer(&strp->rx_msg_timer, strp_rx_msg_timeout,
(unsigned long)strp);
INIT_WORK(&strp->rx_work, strp_rx_work);
strp->cb.rcv_msg = cb->rcv_msg;
strp->cb.parse_msg = cb->parse_msg;
strp->cb.read_sock_done = cb->read_sock_done ? : default_read_sock_done;
strp->cb.abort_parser = cb->abort_parser ? : strp_abort_rx_strp;
return 0;
}
tcp协议栈调用strp->cb流程
sk_data_ready -->strp_read_sock -->sock->ops->read_sock(strp->sk, &desc, strp_recv)
void strp_data_ready(struct strparser *strp)
{
if (unlikely(strp->rx_stopped))
return;
/* This check is needed to synchronize with do_strp_rx_work.
* do_strp_rx_work acquires a process lock (lock_sock) whereas
* the lock held here is bh_lock_sock. The two locks can be
* held by different threads at the same time, but bh_lock_sock
* allows a thread in BH context to safely check if the process
* lock is held. In this case, if the lock is held, queue work.
*/
if (sock_owned_by_user(strp->sk)) {
queue_work(strp_wq, &strp->rx_work);
return;
}
if (strp->rx_paused)
return;
if (strp->rx_need_bytes) {
if (strp_peek_len(strp) >= strp->rx_need_bytes)
strp->rx_need_bytes = 0;
else
return;
}
if (strp_read_sock(strp) == -ENOMEM)
queue_work(strp_wq, &strp->rx_work);
}
static int strp_read_sock(struct strparser *strp)
{
struct socket *sock = strp->sk->sk_socket;
read_descriptor_t desc;
desc.arg.data = strp;
desc.error = 0;
desc.count = 1; /* give more than one skb per call */
/* sk should be locked here, so okay to do read_sock */
sock->ops->read_sock(strp->sk, &desc, strp_recv);
desc.error = strp->cb.read_sock_done(strp, desc.error);
return desc.error;
}
static int strp_recv(read_descriptor_t *desc, struct sk_buff *orig_skb,
unsigned int orig_offset, size_t orig_len)
{
struct strparser *strp = (struct strparser *)desc->arg.data;
return __strp_recv(desc, orig_skb, orig_offset, orig_len,
strp->sk->sk_rcvbuf, strp->sk->sk_rcvtimeo);
}
static int __strp_recv(read_descriptor_t *desc, struct sk_buff *orig_skb,
unsigned int orig_offset, size_t orig_len,
size_t max_msg_size, long timeo)
{
len = (*strp->cb.parse_msg)(strp, head);
/* Give skb to upper layer */
strp->cb.rcv_msg(strp, head);
}
定时器
static void strp_rx_msg_timeout(unsigned long arg)
{
struct strparser *strp = (struct strparser *)arg;
/* Message assembly timed out */
STRP_STATS_INCR(strp->stats.rx_msg_timeouts);
lock_sock(strp->sk);
strp->cb.abort_parser(strp, ETIMEDOUT);
release_sock(strp->sk);
}