summaryrefslogtreecommitdiff
path: root/net/mptcp
diff options
context:
space:
mode:
Diffstat (limited to 'net/mptcp')
-rw-r--r--net/mptcp/options.c39
-rw-r--r--net/mptcp/protocol.c55
2 files changed, 39 insertions, 55 deletions
diff --git a/net/mptcp/options.c b/net/mptcp/options.c
index c41273cefc51..f0f22eb4fd5f 100644
--- a/net/mptcp/options.c
+++ b/net/mptcp/options.c
@@ -485,11 +485,11 @@ static bool mptcp_established_options_mp(struct sock *sk, struct sk_buff *skb,
mpext = mptcp_get_ext(skb);
data_len = mpext ? mpext->data_len : 0;
- /* we will check ext_copy.data_len in mptcp_write_options() to
+ /* we will check ops->data_len in mptcp_write_options() to
* discriminate between TCPOLEN_MPTCP_MPC_ACK_DATA and
* TCPOLEN_MPTCP_MPC_ACK
*/
- opts->ext_copy.data_len = data_len;
+ opts->data_len = data_len;
opts->suboptions = OPTION_MPTCP_MPC_ACK;
opts->sndr_key = subflow->local_key;
opts->rcvr_key = subflow->remote_key;
@@ -505,9 +505,9 @@ static bool mptcp_established_options_mp(struct sock *sk, struct sk_buff *skb,
len = TCPOLEN_MPTCP_MPC_ACK_DATA;
if (opts->csum_reqd) {
/* we need to propagate more info to csum the pseudo hdr */
- opts->ext_copy.data_seq = mpext->data_seq;
- opts->ext_copy.subflow_seq = mpext->subflow_seq;
- opts->ext_copy.csum = mpext->csum;
+ opts->data_seq = mpext->data_seq;
+ opts->subflow_seq = mpext->subflow_seq;
+ opts->csum = mpext->csum;
len += TCPOLEN_MPTCP_DSS_CHECKSUM;
}
*size = ALIGN(len, 4);
@@ -1227,7 +1227,7 @@ static void mptcp_set_rwin(const struct tcp_sock *tp)
WRITE_ONCE(msk->rcv_wnd_sent, ack_seq);
}
-static u16 mptcp_make_csum(const struct mptcp_ext *mpext)
+static u16 __mptcp_make_csum(u64 data_seq, u32 subflow_seq, u16 data_len, __sum16 sum)
{
struct csum_pseudo_header header;
__wsum csum;
@@ -1237,15 +1237,21 @@ static u16 mptcp_make_csum(const struct mptcp_ext *mpext)
* always the 64-bit value, irrespective of what length is used in the
* DSS option itself.
*/
- header.data_seq = cpu_to_be64(mpext->data_seq);
- header.subflow_seq = htonl(mpext->subflow_seq);
- header.data_len = htons(mpext->data_len);
+ header.data_seq = cpu_to_be64(data_seq);
+ header.subflow_seq = htonl(subflow_seq);
+ header.data_len = htons(data_len);
header.csum = 0;
- csum = csum_partial(&header, sizeof(header), ~csum_unfold(mpext->csum));
+ csum = csum_partial(&header, sizeof(header), ~csum_unfold(sum));
return (__force u16)csum_fold(csum);
}
+static u16 mptcp_make_csum(const struct mptcp_ext *mpext)
+{
+ return __mptcp_make_csum(mpext->data_seq, mpext->subflow_seq, mpext->data_len,
+ mpext->csum);
+}
+
void mptcp_write_options(__be32 *ptr, const struct tcp_sock *tp,
struct mptcp_out_options *opts)
{
@@ -1337,7 +1343,7 @@ void mptcp_write_options(__be32 *ptr, const struct tcp_sock *tp,
len = TCPOLEN_MPTCP_MPC_SYN;
} else if (OPTION_MPTCP_MPC_SYNACK & opts->suboptions) {
len = TCPOLEN_MPTCP_MPC_SYNACK;
- } else if (opts->ext_copy.data_len) {
+ } else if (opts->data_len) {
len = TCPOLEN_MPTCP_MPC_ACK_DATA;
if (opts->csum_reqd)
len += TCPOLEN_MPTCP_DSS_CHECKSUM;
@@ -1366,14 +1372,17 @@ void mptcp_write_options(__be32 *ptr, const struct tcp_sock *tp,
put_unaligned_be64(opts->rcvr_key, ptr);
ptr += 2;
- if (!opts->ext_copy.data_len)
+ if (!opts->data_len)
goto mp_capable_done;
if (opts->csum_reqd) {
- put_unaligned_be32(opts->ext_copy.data_len << 16 |
- mptcp_make_csum(&opts->ext_copy), ptr);
+ put_unaligned_be32(opts->data_len << 16 |
+ __mptcp_make_csum(opts->data_seq,
+ opts->subflow_seq,
+ opts->data_len,
+ opts->csum), ptr);
} else {
- put_unaligned_be32(opts->ext_copy.data_len << 16 |
+ put_unaligned_be32(opts->data_len << 16 |
TCPOPT_NOP << 8 | TCPOPT_NOP, ptr);
}
ptr += 1;
diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
index e5df0b5971c8..d073b2111382 100644
--- a/net/mptcp/protocol.c
+++ b/net/mptcp/protocol.c
@@ -528,7 +528,6 @@ static bool mptcp_check_data_fin(struct sock *sk)
sk->sk_shutdown |= RCV_SHUTDOWN;
smp_mb__before_atomic(); /* SHUTDOWN must be visible first */
- set_bit(MPTCP_DATA_READY, &msk->flags);
switch (sk->sk_state) {
case TCP_ESTABLISHED:
@@ -742,10 +741,9 @@ void mptcp_data_ready(struct sock *sk, struct sock *ssk)
/* Wake-up the reader only for in-sequence data */
mptcp_data_lock(sk);
- if (move_skbs_to_msk(msk, ssk)) {
- set_bit(MPTCP_DATA_READY, &msk->flags);
+ if (move_skbs_to_msk(msk, ssk))
sk->sk_data_ready(sk);
- }
+
mptcp_data_unlock(sk);
}
@@ -847,7 +845,6 @@ static void mptcp_check_for_eof(struct mptcp_sock *msk)
sk->sk_shutdown |= RCV_SHUTDOWN;
smp_mb__before_atomic(); /* SHUTDOWN must be visible first */
- set_bit(MPTCP_DATA_READY, &msk->flags);
sk->sk_data_ready(sk);
}
@@ -1759,21 +1756,6 @@ out:
return copied ? : ret;
}
-static void mptcp_wait_data(struct sock *sk, long *timeo)
-{
- DEFINE_WAIT_FUNC(wait, woken_wake_function);
- struct mptcp_sock *msk = mptcp_sk(sk);
-
- add_wait_queue(sk_sleep(sk), &wait);
- sk_set_bit(SOCKWQ_ASYNC_WAITDATA, sk);
-
- sk_wait_event(sk, timeo,
- test_bit(MPTCP_DATA_READY, &msk->flags), &wait);
-
- sk_clear_bit(SOCKWQ_ASYNC_WAITDATA, sk);
- remove_wait_queue(sk_sleep(sk), &wait);
-}
-
static int __mptcp_recvmsg_mskq(struct mptcp_sock *msk,
struct msghdr *msg,
size_t len, int flags,
@@ -2077,19 +2059,7 @@ static int mptcp_recvmsg(struct sock *sk, struct msghdr *msg, size_t len,
}
pr_debug("block timeout %ld", timeo);
- mptcp_wait_data(sk, &timeo);
- }
-
- if (skb_queue_empty_lockless(&sk->sk_receive_queue) &&
- skb_queue_empty(&msk->receive_queue)) {
- /* entire backlog drained, clear DATA_READY. */
- clear_bit(MPTCP_DATA_READY, &msk->flags);
-
- /* .. race-breaker: ssk might have gotten new data
- * after last __mptcp_move_skbs() returned false.
- */
- if (unlikely(__mptcp_move_skbs(msk)))
- set_bit(MPTCP_DATA_READY, &msk->flags);
+ sk_wait_data(sk, &timeo, NULL);
}
out_err:
@@ -2098,9 +2068,9 @@ out_err:
tcp_recv_timestamp(msg, sk, &tss);
}
- pr_debug("msk=%p data_ready=%d rx queue empty=%d copied=%d",
- msk, test_bit(MPTCP_DATA_READY, &msk->flags),
- skb_queue_empty_lockless(&sk->sk_receive_queue), copied);
+ pr_debug("msk=%p rx queue empty=%d:%d copied=%d",
+ msk, skb_queue_empty_lockless(&sk->sk_receive_queue),
+ skb_queue_empty(&msk->receive_queue), copied);
if (!(flags & MSG_PEEK))
mptcp_rcv_space_adjust(msk, copied);
@@ -2368,7 +2338,6 @@ static void mptcp_check_fastclose(struct mptcp_sock *msk)
inet_sk_state_store(sk, TCP_CLOSE);
sk->sk_shutdown = SHUTDOWN_MASK;
smp_mb__before_atomic(); /* SHUTDOWN must be visible first */
- set_bit(MPTCP_DATA_READY, &msk->flags);
set_bit(MPTCP_WORK_CLOSE_SUBFLOW, &msk->flags);
mptcp_close_wake_up(sk);
@@ -3385,8 +3354,14 @@ unlock_fail:
static __poll_t mptcp_check_readable(struct mptcp_sock *msk)
{
- return test_bit(MPTCP_DATA_READY, &msk->flags) ? EPOLLIN | EPOLLRDNORM :
- 0;
+ /* Concurrent splices from sk_receive_queue into receive_queue will
+ * always show at least one non-empty queue when checked in this order.
+ */
+ if (skb_queue_empty_lockless(&((struct sock *)msk)->sk_receive_queue) &&
+ skb_queue_empty_lockless(&msk->receive_queue))
+ return 0;
+
+ return EPOLLIN | EPOLLRDNORM;
}
static __poll_t mptcp_check_writeable(struct mptcp_sock *msk)
@@ -3421,7 +3396,7 @@ static __poll_t mptcp_poll(struct file *file, struct socket *sock,
state = inet_sk_state_load(sk);
pr_debug("msk=%p state=%d flags=%lx", msk, state, msk->flags);
if (state == TCP_LISTEN)
- return mptcp_check_readable(msk);
+ return test_bit(MPTCP_DATA_READY, &msk->flags) ? EPOLLIN | EPOLLRDNORM : 0;
if (state != TCP_SYN_SENT && state != TCP_SYN_RECV) {
mask |= mptcp_check_readable(msk);
generated by cgit v1.2.3 (git 2.46.0) at 2025-08-25 12:36:21 +0000