Commit 29807318 authored Jun 15, 2015 by Neil Brown Committed by Trond Myklebust Jun 16, 2015

SUNRPC: never enqueue a ->rq_cong request on ->sending



If the sending queue has a task without ->rq_cong set at the front,
and then a number of tasks with ->rq_cong set such that they use
the entire congestion window, then the queue deadlocks.  The first
entry cannot be processed until later entries complete.

This scenario has been seen with a client using UDP to access a server,
and the network connection breaking for a period of time - it doesn't
recover.

It never really makes sense for an ->rq_cong request to be on the ->sending
queue, but it can happen when a request is being retried, and finds
the transport if locked (XPRT_LOCKED).  In this case we simple call
__xprt_put_cong() and the deadlock goes away.

Signed-off-by: NeilBrown <neilb@suse.de>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>

parent 6f02dc88

net/sunrpc/xprt.c

+3 −0

Original line number	Diff line number	Diff line
		@@ -68,6 +68,7 @@ static void xprt_init(struct rpc_xprt xprt, struct net net);
		static void xprt_request_init(struct rpc_task , struct rpc_xprt );
		static void xprt_connect_status(struct rpc_task *task);
		static int __xprt_get_cong(struct rpc_xprt , struct rpc_task );
		static void __xprt_put_cong(struct rpc_xprt , struct rpc_rqst );
		static void xprt_destroy(struct rpc_xprt *xprt);

		static DEFINE_SPINLOCK(xprt_list_lock);
		@@ -250,6 +251,8 @@ int xprt_reserve_xprt_cong(struct rpc_xprt xprt, struct rpc_task task)
		}
		xprt_clear_locked(xprt);
		out_sleep:
		if (req)
		__xprt_put_cong(xprt, req);
		dprintk("RPC: %5u failed to lock transport %p\n", task->tk_pid, xprt);
		task->tk_timeout = 0;
		task->tk_status = -EAGAIN;

Admin message