From: Eric Dumazet edumazet@google.com
[ Upstream commit 1f142c17d19a5618d5a633195a46f2c8be9bf232 ]
tcp_memory_pressure is read without holding any lock, and its value could be changed on other cpus.
Use READ_ONCE() to annotate these lockless reads.
The write side is already using atomic ops.
Fixes: b8da51ebb1aa ("tcp: introduce tcp_under_memory_pressure()") Signed-off-by: Eric Dumazet edumazet@google.com Signed-off-by: Jakub Kicinski jakub.kicinski@netronome.com Signed-off-by: Sasha Levin sashal@kernel.org --- include/net/tcp.h | 2 +- net/ipv4/tcp.c | 4 ++-- 2 files changed, 3 insertions(+), 3 deletions(-)
diff --git a/include/net/tcp.h b/include/net/tcp.h index 00d10f0e11949..c963023103149 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -289,7 +289,7 @@ static inline bool tcp_under_memory_pressure(const struct sock *sk) mem_cgroup_under_socket_pressure(sk->sk_memcg)) return true;
- return tcp_memory_pressure; + return READ_ONCE(tcp_memory_pressure); } /* * The next routines deal with comparing 32 bit unsigned ints diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 8f07655718f34..db1eceda2359c 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -328,7 +328,7 @@ void tcp_enter_memory_pressure(struct sock *sk) { unsigned long val;
- if (tcp_memory_pressure) + if (READ_ONCE(tcp_memory_pressure)) return; val = jiffies;
@@ -343,7 +343,7 @@ void tcp_leave_memory_pressure(struct sock *sk) { unsigned long val;
- if (!tcp_memory_pressure) + if (!READ_ONCE(tcp_memory_pressure)) return; val = xchg(&tcp_memory_pressure, 0); if (val)