diff options
| author | Guangguan Wang <guangguan.wang@linux.alibaba.com> | 2024-12-11 12:21:17 +0300 |
|---|---|---|
| committer | Greg Kroah-Hartman <gregkh@linuxfoundation.org> | 2024-12-27 15:58:47 +0300 |
| commit | 2e0786b3006e8c872320e86393920d580b7fba23 (patch) | |
| tree | 45a7d0492b1452a0153e7ffac0967a3342c8e6ca | |
| parent | 2627c3e8646932dfc7b9722c88c2e1ffcf7a9fb2 (diff) | |
| download | linux-2e0786b3006e8c872320e86393920d580b7fba23.tar.xz | |
net/smc: check sndbuf_space again after NOSPACE flag is set in smc_poll
[ Upstream commit 679e9ddcf90dbdf98aaaa71a492454654b627bcb ]
When application sending data more than sndbuf_space, there have chances
application will sleep in epoll_wait, and will never be wakeup again. This
is caused by a race between smc_poll and smc_cdc_tx_handler.
application tasklet
smc_tx_sendmsg(len > sndbuf_space) |
epoll_wait for EPOLL_OUT,timeout=0 |
smc_poll |
if (!smc->conn.sndbuf_space) |
| smc_cdc_tx_handler
| atomic_add sndbuf_space
| smc_tx_sndbuf_nonfull
| if (!test_bit SOCK_NOSPACE)
| do not sk_write_space;
set_bit SOCK_NOSPACE; |
return mask=0; |
Application will sleep in epoll_wait as smc_poll returns 0. And
smc_cdc_tx_handler will not call sk_write_space because the SOCK_NOSPACE
has not be set. If there is no inflight cdc msg, sk_write_space will not be
called any more, and application will sleep in epoll_wait forever.
So check sndbuf_space again after NOSPACE flag is set to break the race.
Fixes: 8dce2786a290 ("net/smc: smc_poll improvements")
Signed-off-by: Guangguan Wang <guangguan.wang@linux.alibaba.com>
Suggested-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>
| -rw-r--r-- | net/smc/af_smc.c | 7 |
1 files changed, 7 insertions, 0 deletions
diff --git a/net/smc/af_smc.c b/net/smc/af_smc.c index 77c6c0dff069..06d607e676f6 100644 --- a/net/smc/af_smc.c +++ b/net/smc/af_smc.c @@ -2888,6 +2888,13 @@ static __poll_t smc_poll(struct file *file, struct socket *sock, } else { sk_set_bit(SOCKWQ_ASYNC_NOSPACE, sk); set_bit(SOCK_NOSPACE, &sk->sk_socket->flags); + + if (sk->sk_state != SMC_INIT) { + /* Race breaker the same way as tcp_poll(). */ + smp_mb__after_atomic(); + if (atomic_read(&smc->conn.sndbuf_space)) + mask |= EPOLLOUT | EPOLLWRNORM; + } } if (atomic_read(&smc->conn.bytes_to_rcv)) mask |= EPOLLIN | EPOLLRDNORM; |