Commit 02c558b2 authored by John Fastabend's avatar John Fastabend Committed by Daniel Borkmann
Browse files

bpf: sockmap, support for msg_peek in sk_msg with redirect ingress 



This adds support for the MSG_PEEK flag when doing redirect to ingress
and receiving on the sk_msg psock queue. Previously the flag was
being ignored which could confuse applications if they expected the
flag to work as normal.

Signed-off-by: default avatarJohn Fastabend <john.fastabend@gmail.com>
Signed-off-by: default avatarDaniel Borkmann <daniel@iogearbox.net>
parent 8734a162

include/net/tcp.h

+1 −1
Original line number Original line Diff line number Diff line
@@ -2089,7 +2089,7 @@ int tcp_bpf_sendmsg_redir(struct sock *sk, struct sk_msg *msg, u32 bytes, 
int tcp_bpf_recvmsg(struct sock *sk, struct msghdr *msg, size_t len,

int tcp_bpf_recvmsg(struct sock *sk, struct msghdr *msg, size_t len,

		    int nonblock, int flags, int *addr_len);

		    int nonblock, int flags, int *addr_len);

int __tcp_bpf_recvmsg(struct sock *sk, struct sk_psock *psock,

int __tcp_bpf_recvmsg(struct sock *sk, struct sk_psock *psock,

		      struct msghdr *msg, int len);

		      struct msghdr *msg, int len, int flags);





/* Call BPF_SOCK_OPS program that returns an int. If the return value

/* Call BPF_SOCK_OPS program that returns an int. If the return value

 * is < 0, then the BPF op failed (for example if the loaded BPF

 * is < 0, then the BPF op failed (for example if the loaded BPF

net/ipv4/tcp_bpf.c

+27 −15
Original line number Original line Diff line number Diff line
@@ -39,17 +39,19 @@ static int tcp_bpf_wait_data(struct sock *sk, struct sk_psock *psock, 
}

}





int __tcp_bpf_recvmsg(struct sock *sk, struct sk_psock *psock,

int __tcp_bpf_recvmsg(struct sock *sk, struct sk_psock *psock,

		      struct msghdr *msg, int len)

		      struct msghdr *msg, int len, int flags)

{

{

	struct iov_iter *iter = &msg->msg_iter;

	struct iov_iter *iter = &msg->msg_iter;

	int peek = flags & MSG_PEEK;

	int i, ret, copied = 0;

	int i, ret, copied = 0;



	while (copied != len) {

		struct scatterlist *sge;

	struct sk_msg *msg_rx;

	struct sk_msg *msg_rx;





	msg_rx = list_first_entry_or_null(&psock->ingress_msg,

	msg_rx = list_first_entry_or_null(&psock->ingress_msg,

					  struct sk_msg, list);

					  struct sk_msg, list);



	while (copied != len) {

		struct scatterlist *sge;



		if (unlikely(!msg_rx))

		if (unlikely(!msg_rx))

			break;

			break;
@@ -70,22 +72,30 @@ int __tcp_bpf_recvmsg(struct sock *sk, struct sk_psock *psock, 
			}

			}





			copied += copy;

			copied += copy;

			if (likely(!peek)) {

				sge->offset += copy;

				sge->offset += copy;

				sge->length -= copy;

				sge->length -= copy;

				sk_mem_uncharge(sk, copy);

				sk_mem_uncharge(sk, copy);

				msg_rx->sg.size -= copy;

				msg_rx->sg.size -= copy;



				if (!sge->length) {

				if (!sge->length) {

				i++;

					sk_msg_iter_var_next(i);

				if (i == MAX_SKB_FRAGS)

					i = 0;

					if (!msg_rx->skb)

					if (!msg_rx->skb)

						put_page(page);

						put_page(page);

				}

				}

			} else {

				sk_msg_iter_var_next(i);

			}





			if (copied == len)

			if (copied == len)

				break;

				break;

		} while (i != msg_rx->sg.end);

		} while (i != msg_rx->sg.end);





		if (unlikely(peek)) {

			msg_rx = list_next_entry(msg_rx, list);

			continue;

		}



		msg_rx->sg.start = i;

		msg_rx->sg.start = i;

		if (!sge->length && msg_rx->sg.start == msg_rx->sg.end) {

		if (!sge->length && msg_rx->sg.start == msg_rx->sg.end) {

			list_del(&msg_rx->list);

			list_del(&msg_rx->list);
@@ -93,6 +103,8 @@ int __tcp_bpf_recvmsg(struct sock *sk, struct sk_psock *psock, 
				consume_skb(msg_rx->skb);

				consume_skb(msg_rx->skb);

			kfree(msg_rx);

			kfree(msg_rx);

		}

		}

		msg_rx = list_first_entry_or_null(&psock->ingress_msg,

						  struct sk_msg, list);

	}

	}





	return copied;

	return copied;
@@ -115,7 +127,7 @@ int tcp_bpf_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, 
		return tcp_recvmsg(sk, msg, len, nonblock, flags, addr_len);

		return tcp_recvmsg(sk, msg, len, nonblock, flags, addr_len);

	lock_sock(sk);

	lock_sock(sk);

msg_bytes_ready:

msg_bytes_ready:

	copied = __tcp_bpf_recvmsg(sk, psock, msg, len);

	copied = __tcp_bpf_recvmsg(sk, psock, msg, len, flags);

	if (!copied) {

	if (!copied) {

		int data, err = 0;

		int data, err = 0;

		long timeo;

		long timeo;

net/tls/tls_sw.c

+2 −1
Original line number Original line Diff line number Diff line
@@ -1478,7 +1478,8 @@ int tls_sw_recvmsg(struct sock *sk, 
		skb = tls_wait_data(sk, psock, flags, timeo, &err);

		skb = tls_wait_data(sk, psock, flags, timeo, &err);

		if (!skb) {

		if (!skb) {

			if (psock) {

			if (psock) {

				int ret = __tcp_bpf_recvmsg(sk, psock, msg, len);

				int ret = __tcp_bpf_recvmsg(sk, psock,

							    msg, len, flags);





				if (ret > 0) {

				if (ret > 0) {

					copied += ret;

					copied += ret;

CRA Git | Maintained and supported by SUSTech CRA and CCSE