xprtrdma: Replace use of xdr_stream_pos in rpcrdma_marshal_req

This is a latent bug. xdr_stream_pos works by subtracting
xdr_stream::nwords from xdr_buf::len. But xdr_stream::nwords is not
initialized by xdr_init_encode().

It works today only because all fields in rpcrdma_req::rl_stream
are initialized to zero by rpcrdma_req_create, making the
subtraction in xdr_stream_pos always a no-op.

I found this issue via code inspection. It was introduced by commit
39f4cd9e99 ("xprtrdma: Harden chunk list encoding against send
buffer overflow"), but the code has changed enough since then that
this fix can't be automatically applied to stable.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>

include/trace/events/rpcrdma.h

@@ -470,13 +470,12 @@ TRACE_DEFINE_ENUM(rpcrdma_replych);
 
 TRACE_EVENT(xprtrdma_marshal,
 	TP_PROTO(
-		const struct rpc_rqst *rqst,
-		unsigned int hdrlen,
+		const struct rpcrdma_req *req,
 		unsigned int rtype,
 		unsigned int wtype
 	),
 
-	TP_ARGS(rqst, hdrlen, rtype, wtype),
+	TP_ARGS(req, rtype, wtype),
 
 	TP_STRUCT__entry(
 		__field(unsigned int, task_id)
@@ -491,10 +490,12 @@ TRACE_EVENT(xprtrdma_marshal,
 	),
 
 	TP_fast_assign(
+		const struct rpc_rqst *rqst = &req->rl_slot;
+
 		__entry->task_id = rqst->rq_task->tk_pid;
 		__entry->client_id = rqst->rq_task->tk_client->cl_clid;
 		__entry->xid = be32_to_cpu(rqst->rq_xid);
-		__entry->hdrlen = hdrlen;
+		__entry->hdrlen = req->rl_hdrbuf.len;
 		__entry->headlen = rqst->rq_snd_buf.head[0].iov_len;
 		__entry->pagelen = rqst->rq_snd_buf.page_len;
 		__entry->taillen = rqst->rq_snd_buf.tail[0].iov_len;