From: Kuniyuki Iwashima kuniyu@amazon.com
stable inclusion from stable-v5.10.134 commit 474510e174fb2bc9751e04885e4cc30023bcf9d7 category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I5ZVR7
Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=...
--------------------------------
[ Upstream commit 46778cd16e6a5ad1b2e3a91f6c057c907379418e ]
While reading sysctl_tcp_reordering, it can be changed concurrently. Thus, we need to add READ_ONCE() to its readers.
Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Signed-off-by: Kuniyuki Iwashima kuniyu@amazon.com Signed-off-by: David S. Miller davem@davemloft.net Signed-off-by: Sasha Levin sashal@kernel.org Signed-off-by: Zheng Zengkai zhengzengkai@huawei.com Reviewed-by: Wei Li liwei391@huawei.com --- net/ipv4/tcp.c | 2 +- net/ipv4/tcp_input.c | 10 +++++++--- net/ipv4/tcp_metrics.c | 3 ++- 3 files changed, 10 insertions(+), 5 deletions(-)
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 1270772cb1f3..e2fdb4f9b66a 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -444,7 +444,7 @@ void tcp_init_sock(struct sock *sk) tp->snd_cwnd_clamp = ~0; tp->mss_cache = TCP_MSS_DEFAULT;
- tp->reordering = sock_net(sk)->ipv4.sysctl_tcp_reordering; + tp->reordering = READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_reordering); tcp_assign_congestion_control(sk);
tp->tsoffset = 0; diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index a751541d48ef..b87d2ceab910 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -2099,6 +2099,7 @@ void tcp_enter_loss(struct sock *sk) struct tcp_sock *tp = tcp_sk(sk); struct net *net = sock_net(sk); bool new_recovery = icsk->icsk_ca_state < TCP_CA_Recovery; + u8 reordering;
tcp_timeout_mark_lost(sk);
@@ -2119,10 +2120,12 @@ void tcp_enter_loss(struct sock *sk) /* Timeout in disordered state after receiving substantial DUPACKs * suggests that the degree of reordering is over-estimated. */ + reordering = READ_ONCE(net->ipv4.sysctl_tcp_reordering); if (icsk->icsk_ca_state <= TCP_CA_Disorder && - tp->sacked_out >= net->ipv4.sysctl_tcp_reordering) + tp->sacked_out >= reordering) tp->reordering = min_t(unsigned int, tp->reordering, - net->ipv4.sysctl_tcp_reordering); + reordering); + tcp_set_ca_state(sk, TCP_CA_Loss); tp->high_seq = tp->snd_nxt; tcp_ecn_queue_cwr(tp); @@ -3411,7 +3414,8 @@ static inline bool tcp_may_raise_cwnd(const struct sock *sk, const int flag) * new SACK or ECE mark may first advance cwnd here and later reduce * cwnd in tcp_fastretrans_alert() based on more states. */ - if (tcp_sk(sk)->reordering > sock_net(sk)->ipv4.sysctl_tcp_reordering) + if (tcp_sk(sk)->reordering > + READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_reordering)) return flag & FLAG_FORWARD_PROGRESS;
return flag & FLAG_DATA_ACKED; diff --git a/net/ipv4/tcp_metrics.c b/net/ipv4/tcp_metrics.c index 6b27c481fe18..8d7e32f4abf6 100644 --- a/net/ipv4/tcp_metrics.c +++ b/net/ipv4/tcp_metrics.c @@ -428,7 +428,8 @@ void tcp_update_metrics(struct sock *sk) if (!tcp_metric_locked(tm, TCP_METRIC_REORDERING)) { val = tcp_metric_get(tm, TCP_METRIC_REORDERING); if (val < tp->reordering && - tp->reordering != net->ipv4.sysctl_tcp_reordering) + tp->reordering != + READ_ONCE(net->ipv4.sysctl_tcp_reordering)) tcp_metric_set(tm, TCP_METRIC_REORDERING, tp->reordering); }