Donate to e Foundation | Murena handsets with /e/OS | Own a part of Murena! Learn more

Commit 15f081ca authored by Trond Myklebust's avatar Trond Myklebust
Browse files

SUNRPC: Avoid an unnecessary task reschedule on ENOTCONN 



If the socket is unconnected, and xprt_transmit() returns ENOTCONN, we
currently give up the lock on the transport channel. Doing so means that
the lock automatically gets assigned to the next task in the xprt->sending
queue, and so that task needs to be woken up to do the actual connect.

The following patch aims to avoid that unnecessary task switch.

Signed-off-by: default avatarTrond Myklebust <Trond.Myklebust@netapp.com>
parent a67d18f8

net/sunrpc/clnt.c

+18 −8
Original line number Original line Diff line number Diff line
@@ -1105,15 +1105,25 @@ static void 
call_transmit_status(struct rpc_task *task)

call_transmit_status(struct rpc_task *task)

{

{

	task->tk_action = call_status;

	task->tk_action = call_status;

	switch (task->tk_status) {

	case -EAGAIN:

		break;

	default:

		xprt_end_transmit(task);

		/*

		/*

	 * Special case: if we've been waiting on the socket's write_space()

		 * Special cases: if we've been waiting on the

	 * callback, then don't call xprt_end_transmit().

		 * socket's write_space() callback, or if the

		 * socket just returned a connection error,

		 * then hold onto the transport lock.

		 */

		 */

	if (task->tk_status == -EAGAIN)

	case -ECONNREFUSED:

		return;

	case -ENOTCONN:

	xprt_end_transmit(task);

	case -EHOSTDOWN:

	case -EHOSTUNREACH:

	case -ENETUNREACH:

		rpc_task_force_reencode(task);

		rpc_task_force_reencode(task);

	}

	}

}





/*

/*

 * 6.	Sort out the RPC call status

 * 6.	Sort out the RPC call status