[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-Id: <20191009195553.154443-1-edumazet@google.com>
Date: Wed, 9 Oct 2019 12:55:53 -0700
From: Eric Dumazet <edumazet@...gle.com>
To: "David S . Miller" <davem@...emloft.net>
Cc: netdev <netdev@...r.kernel.org>,
Eric Dumazet <edumazet@...gle.com>,
Eric Dumazet <eric.dumazet@...il.com>,
Jakub Kicinski <jakub.kicinski@...ronome.com>
Subject: [PATCH net] net: avoid possible false sharing in sk_leave_memory_pressure()
As mentioned in https://github.com/google/ktsan/wiki/READ_ONCE-and-WRITE_ONCE#it-may-improve-performance
a C compiler can legally transform :
if (memory_pressure && *memory_pressure)
*memory_pressure = 0;
to :
if (memory_pressure)
*memory_pressure = 0;
Fixes: 0604475119de ("tcp: add TCPMemoryPressuresChrono counter")
Fixes: 180d8cd942ce ("foundations of per-cgroup memory pressure controlling.")
Fixes: 3ab224be6d69 ("[NET] CORE: Introducing new memory accounting interface.")
Signed-off-by: Eric Dumazet <edumazet@...gle.com>
---
net/core/sock.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/net/core/sock.c b/net/core/sock.c
index fac2b4d80de5e54c77628073d7930bddf8a10cb3..50647a10fdb7f050e963e2734f0d3555fa4bd7aa 100644
--- a/net/core/sock.c
+++ b/net/core/sock.c
@@ -2334,8 +2334,8 @@ static void sk_leave_memory_pressure(struct sock *sk)
} else {
unsigned long *memory_pressure = sk->sk_prot->memory_pressure;
- if (memory_pressure && *memory_pressure)
- *memory_pressure = 0;
+ if (memory_pressure && READ_ONCE(*memory_pressure))
+ WRITE_ONCE(*memory_pressure, 0);
}
}
--
2.23.0.581.g78d2f28ef7-goog
Powered by blists - more mailing lists