4.4-stable review patch. If anyone has any objections, please let me know.
------------------
From: Trond Myklebust <trond.myklebust@xxxxxxxxxxxxxxx>
commit 13331a551ab4df87f7a027d2cab392da96aba1de upstream.
We're seeing hangs in the NFS client code, with loops of the form:
RPC: 30317 xmit incomplete (267368 left of 524448)
RPC: 30317 call_status (status -11)
RPC: 30317 call_transmit (status 0)
RPC: 30317 xprt_prepare_transmit
RPC: 30317 xprt_transmit(524448)
RPC: xs_tcp_send_request(267368) = -11
RPC: 30317 xmit incomplete (267368 left of 524448)
RPC: 30317 call_status (status -11)
RPC: 30317 call_transmit (status 0)
RPC: 30317 xprt_prepare_transmit
RPC: 30317 xprt_transmit(524448)
Turns out commit ceb5d58b2170 ("net: fix sock_wake_async() rcu protection")
moved SOCKWQ_ASYNC_NOSPACE out of sock->flags and into sk->sk_wq->flags,
however it never tried to fix up the code in net/sunrpc.
The new idiom is to use the flags in the RCU protected struct socket_wq.
While we're at it, clear out the now redundant places where we set/clear
SOCKWQ_ASYNC_NOSPACE and SOCK_NOSPACE. In principle, sk_stream_wait_memory()
is supposed to set these for us, so we only need to clear them in the
particular case of our ->write_space() callback.
Fixes: ceb5d58b2170 ("net: fix sock_wake_async() rcu protection")
Cc: Eric Dumazet <edumazet@xxxxxxxxxx>
Signed-off-by: Trond Myklebust <trond.myklebust@xxxxxxxxxxxxxxx>
Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx>
---
net/sunrpc/xprtsock.c | 49 +++++++++++++++++++++----------------------------
1 file changed, 21 insertions(+), 28 deletions(-)
--- a/net/sunrpc/xprtsock.c
+++ b/net/sunrpc/xprtsock.c
@@ -398,7 +398,6 @@ static int xs_sendpages(struct socket *s
if (unlikely(!sock))
return -ENOTSOCK;
- clear_bit(SOCKWQ_ASYNC_NOSPACE, &sock->flags);
if (base != 0) {
addr = NULL;
addrlen = 0;
@@ -442,7 +441,6 @@ static void xs_nospace_callback(struct r
struct sock_xprt *transport = container_of(task->tk_rqstp->rq_xprt, struct sock_xprt, xprt);
transport->inet->sk_write_pending--;
- clear_bit(SOCKWQ_ASYNC_NOSPACE, &transport->sock->flags);
}
/**
@@ -467,20 +465,11 @@ static int xs_nospace(struct rpc_task *t
/* Don't race with disconnect */
if (xprt_connected(xprt)) {
- if (test_bit(SOCKWQ_ASYNC_NOSPACE, &transport->sock->flags)) {
- /*
- * Notify TCP that we're limited by the application
- * window size
- */
- set_bit(SOCK_NOSPACE, &transport->sock->flags);
- sk->sk_write_pending++;
- /* ...and wait for more buffer space */
- xprt_wait_for_buffer_space(task, xs_nospace_callback);
- }
- } else {
- clear_bit(SOCKWQ_ASYNC_NOSPACE, &transport->sock->flags);
+ /* wait for more buffer space */
+ sk->sk_write_pending++;
+ xprt_wait_for_buffer_space(task, xs_nospace_callback);
+ } else
ret = -ENOTCONN;
- }
spin_unlock_bh(&xprt->transport_lock);
@@ -616,9 +605,6 @@ process_status:
case -EAGAIN:
status = xs_nospace(task);
break;
- default:
- dprintk("RPC: sendmsg returned unrecognized error %d\n",
- -status);
case -ENETUNREACH:
case -ENOBUFS:
case -EPIPE:
@@ -626,7 +612,10 @@ process_status:
case -EPERM:
/* When the server has died, an ICMP port unreachable message
* prompts ECONNREFUSED. */
- clear_bit(SOCKWQ_ASYNC_NOSPACE, &transport->sock->flags);
+ break;
+ default:
+ dprintk("RPC: sendmsg returned unrecognized error %d\n",
+ -status);
}
return status;
@@ -706,16 +695,16 @@ static int xs_tcp_send_request(struct rp
case -EAGAIN:
status = xs_nospace(task);
break;
- default:
- dprintk("RPC: sendmsg returned unrecognized error %d\n",
- -status);
case -ECONNRESET:
case -ECONNREFUSED:
case -ENOTCONN:
case -EADDRINUSE:
case -ENOBUFS:
case -EPIPE:
- clear_bit(SOCKWQ_ASYNC_NOSPACE, &transport->sock->flags);
+ break;
+ default:
+ dprintk("RPC: sendmsg returned unrecognized error %d\n",
+ -status);
}
return status;
@@ -1609,19 +1598,23 @@ static void xs_tcp_state_change(struct s
static void xs_write_space(struct sock *sk)
{
- struct socket *sock;
+ struct socket_wq *wq;
struct rpc_xprt *xprt;
- if (unlikely(!(sock = sk->sk_socket)))
+ if (!sk->sk_socket)
return;
- clear_bit(SOCK_NOSPACE, &sock->flags);
+ clear_bit(SOCK_NOSPACE, &sk->sk_socket->flags);
if (unlikely(!(xprt = xprt_from_sock(sk))))
return;
- if (test_and_clear_bit(SOCKWQ_ASYNC_NOSPACE, &sock->flags) == 0)
- return;
+ rcu_read_lock();
+ wq = rcu_dereference(sk->sk_wq);
+ if (!wq || test_and_clear_bit(SOCKWQ_ASYNC_NOSPACE, &wq->flags) == 0)
+ goto out;
xprt_write_space(xprt);
+out:
+ rcu_read_unlock();
}
/**
--
To unsubscribe from this list: send the line "unsubscribe stable" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html