Commit f99fa508 authored by Trond Myklebust's avatar Trond Myklebust Committed by Anna Schumaker
Browse files

SUNRPC/xprtrdma: Fix reconnection locking 



The xprtrdma client code currently relies on the task that initiated the
connect to hold the XPRT_LOCK for the duration of the connection
attempt. If the task is woken early, due to some other event, then that
lock could get released early.
Avoid races by using the same mechanism that the socket code uses of
transferring lock ownership to the RDMA connect worker itself. That
frees us to call rpcrdma_xprt_disconnect() directly since we're now
guaranteed exclusion w.r.t. other callers.

Fixes: 4cf44be6 ("xprtrdma: Fix recursion into rpcrdma_xprt_disconnect()")
Signed-off-by: default avatarTrond Myklebust <trond.myklebust@hammerspace.com>
Signed-off-by: default avatarAnna Schumaker <Anna.Schumaker@Netapp.com>
parent e26d9972

net/sunrpc/xprt.c

+2 −0
Original line number Diff line number Diff line
@@ -877,6 +877,7 @@ bool xprt_lock_connect(struct rpc_xprt *xprt, 
	spin_unlock(&xprt->transport_lock);

	return ret;

}

EXPORT_SYMBOL_GPL(xprt_lock_connect);



void xprt_unlock_connect(struct rpc_xprt *xprt, void *cookie)

{
@@ -893,6 +894,7 @@ void xprt_unlock_connect(struct rpc_xprt *xprt, void *cookie) 
	spin_unlock(&xprt->transport_lock);

	wake_up_bit(&xprt->state, XPRT_LOCKED);

}

EXPORT_SYMBOL_GPL(xprt_unlock_connect);



/**

 * xprt_connect - schedule a transport connect operation

net/sunrpc/xprtrdma/transport.c

+5 −6
Original line number Diff line number Diff line
@@ -250,12 +250,9 @@ xprt_rdma_connect_worker(struct work_struct *work) 
					   xprt->stat.connect_start;

		xprt_set_connected(xprt);

		rc = -EAGAIN;

	} else {

		/* Force a call to xprt_rdma_close to clean up */

		spin_lock(&xprt->transport_lock);

		set_bit(XPRT_CLOSE_WAIT, &xprt->state);

		spin_unlock(&xprt->transport_lock);

	}

	} else

		rpcrdma_xprt_disconnect(r_xprt);

	xprt_unlock_connect(xprt, r_xprt);

	xprt_wake_pending_tasks(xprt, rc);

}
@@ -489,6 +486,8 @@ xprt_rdma_connect(struct rpc_xprt *xprt, struct rpc_task *task) 
	struct rpcrdma_ep *ep = r_xprt->rx_ep;

	unsigned long delay;



	WARN_ON_ONCE(!xprt_lock_connect(xprt, task, r_xprt));



	delay = 0;

	if (ep && ep->re_connect_status != 0) {

		delay = xprt_reconnect_delay(xprt);