From: Paolo Abeni <pab...@redhat.com>

So that we keep per unacked sequence number consistent; since
we update per msk data, use an atomic64 cmpxcgh() to protect
against concurrent updates from multiple subflows.

Initialize the snd_una at connect()/accept() time.

Signed-off-by: Paolo Abeni <pab...@redhat.com>
---
 net/mptcp/options.c  | 45 ++++++++++++++++++++++++++++++++++++++------
 net/mptcp/protocol.c |  2 ++
 net/mptcp/protocol.h |  1 +
 3 files changed, 42 insertions(+), 6 deletions(-)

diff --git a/net/mptcp/options.c b/net/mptcp/options.c
index ce298ecc64f5..2427fff98091 100644
--- a/net/mptcp/options.c
+++ b/net/mptcp/options.c
@@ -540,6 +540,39 @@ bool mptcp_synack_options(const struct request_sock *req, 
unsigned int *size,
        return false;
 }
 
+static u64 expand_ack(u64 old_ack, u64 cur_ack, bool use_64bit)
+{
+       u32 old_ack32, cur_ack32;
+
+       if (use_64bit)
+               return cur_ack;
+
+       old_ack32 = (u32)old_ack;
+       cur_ack32 = (u32)cur_ack;
+       cur_ack = (old_ack & GENMASK_ULL(63, 32)) + cur_ack32;
+       if (unlikely(before(cur_ack32, old_ack32)))
+               return cur_ack + (1LL << 32);
+       return cur_ack;
+}
+
+void update_una(struct mptcp_sock *msk, struct mptcp_options_received *mp_opt)
+{
+       u64 new_snd_una, snd_una, old_snd_una = atomic64_read(&msk->snd_una);
+
+       /* avoid ack expansion on update conflict, to reduce the risk of
+        * wrongly expanding to a future ack sequence number, which is way
+        * more dangerous than missing an ack
+        */
+       new_snd_una = expand_ack(old_snd_una, mp_opt->data_ack, mp_opt->ack64);
+       while (after64(new_snd_una, old_snd_una)) {
+               snd_una = old_snd_una;
+               old_snd_una = atomic64_cmpxchg(&msk->snd_una, snd_una,
+                                              new_snd_una);
+               if (old_snd_una == snd_una)
+                       break;
+       }
+}
+
 void mptcp_incoming_options(struct sock *sk, struct sk_buff *skb,
                            struct tcp_options_received *opt_rx)
 {
@@ -563,6 +596,12 @@ void mptcp_incoming_options(struct sock *sk, struct 
sk_buff *skb,
        if (!mp_opt->dss)
                return;
 
+       /* we can't wait for recvmsg() to update the ack_seq, otherwise
+        * monodirectional flows will stuck
+        */
+       if (msk && mp_opt->use_ack)
+               update_una(msk, mp_opt);
+
        mpext = skb_ext_add(skb, SKB_EXT_MPTCP);
        if (!mpext)
                return;
@@ -579,12 +618,6 @@ void mptcp_incoming_options(struct sock *sk, struct 
sk_buff *skb,
                mpext->use_checksum = mp_opt->use_checksum;
        }
 
-       if (mp_opt->use_ack) {
-               mpext->data_ack = mp_opt->data_ack;
-               mpext->use_ack = 1;
-               mpext->ack64 = mp_opt->ack64;
-       }
-
        mpext->data_fin = mp_opt->data_fin;
 
        if (msk)
diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
index c8ee20963887..2c64435aedd8 100644
--- a/net/mptcp/protocol.c
+++ b/net/mptcp/protocol.c
@@ -688,6 +688,7 @@ static struct sock *mptcp_accept(struct sock *sk, int 
flags, int *err,
 
                mptcp_crypto_key_sha1(msk->remote_key, NULL, &ack_seq);
                msk->write_seq = subflow->idsn + 1;
+               atomic64_set(&msk->snd_una, msk->write_seq);
                ack_seq++;
                msk->ack_seq = ack_seq;
                subflow->map_seq = ack_seq;
@@ -822,6 +823,7 @@ void mptcp_finish_connect(struct sock *sk, int mp_capable)
 
                mptcp_crypto_key_sha1(msk->remote_key, NULL, &ack_seq);
                msk->write_seq = subflow->idsn + 1;
+               atomic64_set(&msk->snd_una, msk->write_seq);
                ack_seq++;
                msk->ack_seq = ack_seq;
                subflow->map_seq = ack_seq;
diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h
index 56df4f46f313..45646e38aa4c 100644
--- a/net/mptcp/protocol.h
+++ b/net/mptcp/protocol.h
@@ -119,6 +119,7 @@ struct mptcp_sock {
        u64             remote_key;
        u64             write_seq;
        u64             ack_seq;
+       atomic64_t      snd_una;
        u32             token;
        unsigned long   flags;
        u16             dport;
-- 
2.23.0

Reply via email to