xref: /freebsd/sys/rpc/clnt_dg.c (revision 123af6ec70016f5556da5972d4d63c7d175c06d3)
1 /*	$NetBSD: clnt_dg.c,v 1.4 2000/07/14 08:40:41 fvdl Exp $	*/
2 
3 /*-
4  * SPDX-License-Identifier: BSD-3-Clause
5  *
6  * Copyright (c) 2009, Sun Microsystems, Inc.
7  * All rights reserved.
8  *
9  * Redistribution and use in source and binary forms, with or without
10  * modification, are permitted provided that the following conditions are met:
11  * - Redistributions of source code must retain the above copyright notice,
12  *   this list of conditions and the following disclaimer.
13  * - Redistributions in binary form must reproduce the above copyright notice,
14  *   this list of conditions and the following disclaimer in the documentation
15  *   and/or other materials provided with the distribution.
16  * - Neither the name of Sun Microsystems, Inc. nor the names of its
17  *   contributors may be used to endorse or promote products derived
18  *   from this software without specific prior written permission.
19  *
20  * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
21  * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
22  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
23  * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE
24  * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
25  * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
26  * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
27  * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
28  * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
29  * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
30  * POSSIBILITY OF SUCH DAMAGE.
31  */
32 /*
33  * Copyright (c) 1986-1991 by Sun Microsystems Inc.
34  */
35 
36 #if defined(LIBC_SCCS) && !defined(lint)
37 #ident	"@(#)clnt_dg.c	1.23	94/04/22 SMI"
38 static char sccsid[] = "@(#)clnt_dg.c 1.19 89/03/16 Copyr 1988 Sun Micro";
39 #endif
40 #include <sys/cdefs.h>
41 __FBSDID("$FreeBSD$");
42 
43 /*
44  * Implements a connectionless client side RPC.
45  */
46 
47 #include <sys/param.h>
48 #include <sys/systm.h>
49 #include <sys/kernel.h>
50 #include <sys/lock.h>
51 #include <sys/malloc.h>
52 #include <sys/mbuf.h>
53 #include <sys/mutex.h>
54 #include <sys/pcpu.h>
55 #include <sys/proc.h>
56 #include <sys/socket.h>
57 #include <sys/socketvar.h>
58 #include <sys/time.h>
59 #include <sys/uio.h>
60 
61 #include <net/vnet.h>
62 
63 #include <rpc/rpc.h>
64 #include <rpc/rpc_com.h>
65 
66 
67 #ifdef _FREEFALL_CONFIG
68 /*
69  * Disable RPC exponential back-off for FreeBSD.org systems.
70  */
71 #define	RPC_MAX_BACKOFF		1 /* second */
72 #else
73 #define	RPC_MAX_BACKOFF		30 /* seconds */
74 #endif
75 
76 static bool_t time_not_ok(struct timeval *);
77 static enum clnt_stat clnt_dg_call(CLIENT *, struct rpc_callextra *,
78     rpcproc_t, struct mbuf *, struct mbuf **, struct timeval);
79 static void clnt_dg_geterr(CLIENT *, struct rpc_err *);
80 static bool_t clnt_dg_freeres(CLIENT *, xdrproc_t, void *);
81 static void clnt_dg_abort(CLIENT *);
82 static bool_t clnt_dg_control(CLIENT *, u_int, void *);
83 static void clnt_dg_close(CLIENT *);
84 static void clnt_dg_destroy(CLIENT *);
85 static int clnt_dg_soupcall(struct socket *so, void *arg, int waitflag);
86 
87 static struct clnt_ops clnt_dg_ops = {
88 	.cl_call =	clnt_dg_call,
89 	.cl_abort =	clnt_dg_abort,
90 	.cl_geterr =	clnt_dg_geterr,
91 	.cl_freeres =	clnt_dg_freeres,
92 	.cl_close =	clnt_dg_close,
93 	.cl_destroy =	clnt_dg_destroy,
94 	.cl_control =	clnt_dg_control
95 };
96 
97 /*
98  * A pending RPC request which awaits a reply. Requests which have
99  * received their reply will have cr_xid set to zero and cr_mrep to
100  * the mbuf chain of the reply.
101  */
102 struct cu_request {
103 	TAILQ_ENTRY(cu_request) cr_link;
104 	CLIENT			*cr_client;	/* owner */
105 	uint32_t		cr_xid;		/* XID of request */
106 	struct mbuf		*cr_mrep;	/* reply received by upcall */
107 	int			cr_error;	/* any error from upcall */
108 	char			cr_verf[MAX_AUTH_BYTES]; /* reply verf */
109 };
110 
111 TAILQ_HEAD(cu_request_list, cu_request);
112 
113 #define MCALL_MSG_SIZE 24
114 
115 /*
116  * This structure is pointed to by the socket buffer's sb_upcallarg
117  * member. It is separate from the client private data to facilitate
118  * multiple clients sharing the same socket. The cs_lock mutex is used
119  * to protect all fields of this structure, the socket's receive
120  * buffer SOCKBUF_LOCK is used to ensure that exactly one of these
121  * structures is installed on the socket.
122  */
123 struct cu_socket {
124 	struct mtx		cs_lock;
125 	int			cs_refs;	/* Count of clients */
126 	struct cu_request_list	cs_pending;	/* Requests awaiting replies */
127 	int			cs_upcallrefs;	/* Refcnt of upcalls in prog.*/
128 };
129 
130 static void clnt_dg_upcallsdone(struct socket *, struct cu_socket *);
131 
132 /*
133  * Private data kept per client handle
134  */
135 struct cu_data {
136 	int			cu_threads;	/* # threads in clnt_vc_call */
137 	bool_t			cu_closing;	/* TRUE if we are closing */
138 	bool_t			cu_closed;	/* TRUE if we are closed */
139 	struct socket		*cu_socket;	/* connection socket */
140 	bool_t			cu_closeit;	/* opened by library */
141 	struct sockaddr_storage	cu_raddr;	/* remote address */
142 	int			cu_rlen;
143 	struct timeval		cu_wait;	/* retransmit interval */
144 	struct timeval		cu_total;	/* total time for the call */
145 	struct rpc_err		cu_error;
146 	uint32_t		cu_xid;
147 	char			cu_mcallc[MCALL_MSG_SIZE]; /* marshalled callmsg */
148 	size_t			cu_mcalllen;
149 	size_t			cu_sendsz;	/* send size */
150 	size_t			cu_recvsz;	/* recv size */
151 	int			cu_async;
152 	int			cu_connect;	/* Use connect(). */
153 	int			cu_connected;	/* Have done connect(). */
154 	const char		*cu_waitchan;
155 	int			cu_waitflag;
156 	int			cu_cwnd;	/* congestion window */
157 	int			cu_sent;	/* number of in-flight RPCs */
158 	bool_t			cu_cwnd_wait;
159 };
160 
161 #define CWNDSCALE	256
162 #define MAXCWND		(32 * CWNDSCALE)
163 
164 /*
165  * Connection less client creation returns with client handle parameters.
166  * Default options are set, which the user can change using clnt_control().
167  * fd should be open and bound.
168  * NB: The rpch->cl_auth is initialized to null authentication.
169  * 	Caller may wish to set this something more useful.
170  *
171  * sendsz and recvsz are the maximum allowable packet sizes that can be
172  * sent and received. Normally they are the same, but they can be
173  * changed to improve the program efficiency and buffer allocation.
174  * If they are 0, use the transport default.
175  *
176  * If svcaddr is NULL, returns NULL.
177  */
178 CLIENT *
179 clnt_dg_create(
180 	struct socket *so,
181 	struct sockaddr *svcaddr,	/* servers address */
182 	rpcprog_t program,		/* program number */
183 	rpcvers_t version,		/* version number */
184 	size_t sendsz,			/* buffer recv size */
185 	size_t recvsz)			/* buffer send size */
186 {
187 	CLIENT *cl = NULL;		/* client handle */
188 	struct cu_data *cu = NULL;	/* private data */
189 	struct cu_socket *cs = NULL;
190 	struct sockbuf *sb;
191 	struct timeval now;
192 	struct rpc_msg call_msg;
193 	struct __rpc_sockinfo si;
194 	XDR xdrs;
195 	int error;
196 
197 	if (svcaddr == NULL) {
198 		rpc_createerr.cf_stat = RPC_UNKNOWNADDR;
199 		return (NULL);
200 	}
201 
202 	if (!__rpc_socket2sockinfo(so, &si)) {
203 		rpc_createerr.cf_stat = RPC_TLIERROR;
204 		rpc_createerr.cf_error.re_errno = 0;
205 		return (NULL);
206 	}
207 
208 	/*
209 	 * Find the receive and the send size
210 	 */
211 	sendsz = __rpc_get_t_size(si.si_af, si.si_proto, (int)sendsz);
212 	recvsz = __rpc_get_t_size(si.si_af, si.si_proto, (int)recvsz);
213 	if ((sendsz == 0) || (recvsz == 0)) {
214 		rpc_createerr.cf_stat = RPC_TLIERROR; /* XXX */
215 		rpc_createerr.cf_error.re_errno = 0;
216 		return (NULL);
217 	}
218 
219 	cl = mem_alloc(sizeof (CLIENT));
220 
221 	/*
222 	 * Should be multiple of 4 for XDR.
223 	 */
224 	sendsz = rounddown(sendsz + 3, 4);
225 	recvsz = rounddown(recvsz + 3, 4);
226 	cu = mem_alloc(sizeof (*cu));
227 	cu->cu_threads = 0;
228 	cu->cu_closing = FALSE;
229 	cu->cu_closed = FALSE;
230 	(void) memcpy(&cu->cu_raddr, svcaddr, (size_t)svcaddr->sa_len);
231 	cu->cu_rlen = svcaddr->sa_len;
232 	/* Other values can also be set through clnt_control() */
233 	cu->cu_wait.tv_sec = 3;	/* heuristically chosen */
234 	cu->cu_wait.tv_usec = 0;
235 	cu->cu_total.tv_sec = -1;
236 	cu->cu_total.tv_usec = -1;
237 	cu->cu_sendsz = sendsz;
238 	cu->cu_recvsz = recvsz;
239 	cu->cu_async = FALSE;
240 	cu->cu_connect = FALSE;
241 	cu->cu_connected = FALSE;
242 	cu->cu_waitchan = "rpcrecv";
243 	cu->cu_waitflag = 0;
244 	cu->cu_cwnd = MAXCWND / 2;
245 	cu->cu_sent = 0;
246 	cu->cu_cwnd_wait = FALSE;
247 	(void) getmicrotime(&now);
248 	cu->cu_xid = __RPC_GETXID(&now);
249 	call_msg.rm_xid = cu->cu_xid;
250 	call_msg.rm_call.cb_prog = program;
251 	call_msg.rm_call.cb_vers = version;
252 	xdrmem_create(&xdrs, cu->cu_mcallc, MCALL_MSG_SIZE, XDR_ENCODE);
253 	if (! xdr_callhdr(&xdrs, &call_msg)) {
254 		rpc_createerr.cf_stat = RPC_CANTENCODEARGS;  /* XXX */
255 		rpc_createerr.cf_error.re_errno = 0;
256 		goto err2;
257 	}
258 	cu->cu_mcalllen = XDR_GETPOS(&xdrs);
259 
260 	/*
261 	 * By default, closeit is always FALSE. It is users responsibility
262 	 * to do a close on it, else the user may use clnt_control
263 	 * to let clnt_destroy do it for him/her.
264 	 */
265 	cu->cu_closeit = FALSE;
266 	cu->cu_socket = so;
267 	error = soreserve(so, (u_long)sendsz, (u_long)recvsz);
268 	if (error != 0) {
269 		rpc_createerr.cf_stat = RPC_FAILED;
270 		rpc_createerr.cf_error.re_errno = error;
271 		goto err2;
272 	}
273 
274 	sb = &so->so_rcv;
275 	SOCKBUF_LOCK(&so->so_rcv);
276 recheck_socket:
277 	if (sb->sb_upcall) {
278 		if (sb->sb_upcall != clnt_dg_soupcall) {
279 			SOCKBUF_UNLOCK(&so->so_rcv);
280 			printf("clnt_dg_create(): socket already has an incompatible upcall\n");
281 			goto err2;
282 		}
283 		cs = (struct cu_socket *) sb->sb_upcallarg;
284 		mtx_lock(&cs->cs_lock);
285 		cs->cs_refs++;
286 		mtx_unlock(&cs->cs_lock);
287 	} else {
288 		/*
289 		 * We are the first on this socket - allocate the
290 		 * structure and install it in the socket.
291 		 */
292 		SOCKBUF_UNLOCK(&so->so_rcv);
293 		cs = mem_alloc(sizeof(*cs));
294 		SOCKBUF_LOCK(&so->so_rcv);
295 		if (sb->sb_upcall) {
296 			/*
297 			 * We have lost a race with some other client.
298 			 */
299 			mem_free(cs, sizeof(*cs));
300 			goto recheck_socket;
301 		}
302 		mtx_init(&cs->cs_lock, "cs->cs_lock", NULL, MTX_DEF);
303 		cs->cs_refs = 1;
304 		cs->cs_upcallrefs = 0;
305 		TAILQ_INIT(&cs->cs_pending);
306 		soupcall_set(so, SO_RCV, clnt_dg_soupcall, cs);
307 	}
308 	SOCKBUF_UNLOCK(&so->so_rcv);
309 
310 	cl->cl_refs = 1;
311 	cl->cl_ops = &clnt_dg_ops;
312 	cl->cl_private = (caddr_t)(void *)cu;
313 	cl->cl_auth = authnone_create();
314 	cl->cl_tp = NULL;
315 	cl->cl_netid = NULL;
316 	return (cl);
317 err2:
318 	mem_free(cl, sizeof (CLIENT));
319 	mem_free(cu, sizeof (*cu));
320 
321 	return (NULL);
322 }
323 
324 static enum clnt_stat
325 clnt_dg_call(
326 	CLIENT		*cl,		/* client handle */
327 	struct rpc_callextra *ext,	/* call metadata */
328 	rpcproc_t	proc,		/* procedure number */
329 	struct mbuf	*args,		/* pointer to args */
330 	struct mbuf	**resultsp,	/* pointer to results */
331 	struct timeval	utimeout)	/* seconds to wait before giving up */
332 {
333 	struct cu_data *cu = (struct cu_data *)cl->cl_private;
334 	struct cu_socket *cs;
335 	struct rpc_timers *rt;
336 	AUTH *auth;
337 	struct rpc_err *errp;
338 	enum clnt_stat stat;
339 	XDR xdrs;
340 	struct rpc_msg reply_msg;
341 	bool_t ok;
342 	int retrans;			/* number of re-transmits so far */
343 	int nrefreshes = 2;		/* number of times to refresh cred */
344 	struct timeval *tvp;
345 	int timeout;
346 	int retransmit_time;
347 	int next_sendtime, starttime, rtt, time_waited, tv = 0;
348 	struct sockaddr *sa;
349 	uint32_t xid = 0;
350 	struct mbuf *mreq = NULL, *results;
351 	struct cu_request *cr;
352 	int error;
353 
354 	cs = cu->cu_socket->so_rcv.sb_upcallarg;
355 	cr = malloc(sizeof(struct cu_request), M_RPC, M_WAITOK);
356 
357 	mtx_lock(&cs->cs_lock);
358 
359 	if (cu->cu_closing || cu->cu_closed) {
360 		mtx_unlock(&cs->cs_lock);
361 		free(cr, M_RPC);
362 		return (RPC_CANTSEND);
363 	}
364 	cu->cu_threads++;
365 
366 	if (ext) {
367 		auth = ext->rc_auth;
368 		errp = &ext->rc_err;
369 	} else {
370 		auth = cl->cl_auth;
371 		errp = &cu->cu_error;
372 	}
373 
374 	cr->cr_client = cl;
375 	cr->cr_mrep = NULL;
376 	cr->cr_error = 0;
377 
378 	if (cu->cu_total.tv_usec == -1) {
379 		tvp = &utimeout; /* use supplied timeout */
380 	} else {
381 		tvp = &cu->cu_total; /* use default timeout */
382 	}
383 	if (tvp->tv_sec || tvp->tv_usec)
384 		timeout = tvtohz(tvp);
385 	else
386 		timeout = 0;
387 
388 	if (cu->cu_connect && !cu->cu_connected) {
389 		mtx_unlock(&cs->cs_lock);
390 		error = soconnect(cu->cu_socket,
391 		    (struct sockaddr *)&cu->cu_raddr, curthread);
392 		mtx_lock(&cs->cs_lock);
393 		if (error) {
394 			errp->re_errno = error;
395 			errp->re_status = stat = RPC_CANTSEND;
396 			goto out;
397 		}
398 		cu->cu_connected = 1;
399 	}
400 	if (cu->cu_connected)
401 		sa = NULL;
402 	else
403 		sa = (struct sockaddr *)&cu->cu_raddr;
404 	time_waited = 0;
405 	retrans = 0;
406 	if (ext && ext->rc_timers) {
407 		rt = ext->rc_timers;
408 		if (!rt->rt_rtxcur)
409 			rt->rt_rtxcur = tvtohz(&cu->cu_wait);
410 		retransmit_time = next_sendtime = rt->rt_rtxcur;
411 	} else {
412 		rt = NULL;
413 		retransmit_time = next_sendtime = tvtohz(&cu->cu_wait);
414 	}
415 
416 	starttime = ticks;
417 
418 call_again:
419 	mtx_assert(&cs->cs_lock, MA_OWNED);
420 
421 	cu->cu_xid++;
422 	xid = cu->cu_xid;
423 
424 send_again:
425 	mtx_unlock(&cs->cs_lock);
426 
427 	mreq = m_gethdr(M_WAITOK, MT_DATA);
428 	KASSERT(cu->cu_mcalllen <= MHLEN, ("RPC header too big"));
429 	bcopy(cu->cu_mcallc, mreq->m_data, cu->cu_mcalllen);
430 	mreq->m_len = cu->cu_mcalllen;
431 
432 	/*
433 	 * The XID is the first thing in the request.
434 	 */
435 	*mtod(mreq, uint32_t *) = htonl(xid);
436 
437 	xdrmbuf_create(&xdrs, mreq, XDR_ENCODE);
438 
439 	if (cu->cu_async == TRUE && args == NULL)
440 		goto get_reply;
441 
442 	if ((! XDR_PUTINT32(&xdrs, &proc)) ||
443 	    (! AUTH_MARSHALL(auth, xid, &xdrs,
444 		m_copym(args, 0, M_COPYALL, M_WAITOK)))) {
445 		errp->re_status = stat = RPC_CANTENCODEARGS;
446 		mtx_lock(&cs->cs_lock);
447 		goto out;
448 	}
449 	mreq->m_pkthdr.len = m_length(mreq, NULL);
450 
451 	cr->cr_xid = xid;
452 	mtx_lock(&cs->cs_lock);
453 
454 	/*
455 	 * Try to get a place in the congestion window.
456 	 */
457 	while (cu->cu_sent >= cu->cu_cwnd) {
458 		cu->cu_cwnd_wait = TRUE;
459 		error = msleep(&cu->cu_cwnd_wait, &cs->cs_lock,
460 		    cu->cu_waitflag, "rpccwnd", 0);
461 		if (error) {
462 			errp->re_errno = error;
463 			if (error == EINTR || error == ERESTART)
464 				errp->re_status = stat = RPC_INTR;
465 			else
466 				errp->re_status = stat = RPC_CANTSEND;
467 			goto out;
468 		}
469 	}
470 	cu->cu_sent += CWNDSCALE;
471 
472 	TAILQ_INSERT_TAIL(&cs->cs_pending, cr, cr_link);
473 	mtx_unlock(&cs->cs_lock);
474 
475 	/*
476 	 * sosend consumes mreq.
477 	 */
478 	error = sosend(cu->cu_socket, sa, NULL, mreq, NULL, 0, curthread);
479 	mreq = NULL;
480 
481 	/*
482 	 * sub-optimal code appears here because we have
483 	 * some clock time to spare while the packets are in flight.
484 	 * (We assume that this is actually only executed once.)
485 	 */
486 	reply_msg.acpted_rply.ar_verf.oa_flavor = AUTH_NULL;
487 	reply_msg.acpted_rply.ar_verf.oa_base = cr->cr_verf;
488 	reply_msg.acpted_rply.ar_verf.oa_length = 0;
489 	reply_msg.acpted_rply.ar_results.where = NULL;
490 	reply_msg.acpted_rply.ar_results.proc = (xdrproc_t)xdr_void;
491 
492 	mtx_lock(&cs->cs_lock);
493 	if (error) {
494 		TAILQ_REMOVE(&cs->cs_pending, cr, cr_link);
495 		errp->re_errno = error;
496 		errp->re_status = stat = RPC_CANTSEND;
497 		cu->cu_sent -= CWNDSCALE;
498 		if (cu->cu_cwnd_wait) {
499 			cu->cu_cwnd_wait = FALSE;
500 			wakeup(&cu->cu_cwnd_wait);
501 		}
502 		goto out;
503 	}
504 
505 	/*
506 	 * Check to see if we got an upcall while waiting for the
507 	 * lock.
508 	 */
509 	if (cr->cr_error) {
510 		TAILQ_REMOVE(&cs->cs_pending, cr, cr_link);
511 		errp->re_errno = cr->cr_error;
512 		errp->re_status = stat = RPC_CANTRECV;
513 		cu->cu_sent -= CWNDSCALE;
514 		if (cu->cu_cwnd_wait) {
515 			cu->cu_cwnd_wait = FALSE;
516 			wakeup(&cu->cu_cwnd_wait);
517 		}
518 		goto out;
519 	}
520 	if (cr->cr_mrep) {
521 		TAILQ_REMOVE(&cs->cs_pending, cr, cr_link);
522 		cu->cu_sent -= CWNDSCALE;
523 		if (cu->cu_cwnd_wait) {
524 			cu->cu_cwnd_wait = FALSE;
525 			wakeup(&cu->cu_cwnd_wait);
526 		}
527 		goto got_reply;
528 	}
529 
530 	/*
531 	 * Hack to provide rpc-based message passing
532 	 */
533 	if (timeout == 0) {
534 		TAILQ_REMOVE(&cs->cs_pending, cr, cr_link);
535 		errp->re_status = stat = RPC_TIMEDOUT;
536 		cu->cu_sent -= CWNDSCALE;
537 		if (cu->cu_cwnd_wait) {
538 			cu->cu_cwnd_wait = FALSE;
539 			wakeup(&cu->cu_cwnd_wait);
540 		}
541 		goto out;
542 	}
543 
544 get_reply:
545 	for (;;) {
546 		/* Decide how long to wait. */
547 		if (next_sendtime < timeout)
548 			tv = next_sendtime;
549 		else
550 			tv = timeout;
551 		tv -= time_waited;
552 
553 		if (tv > 0) {
554 			if (cu->cu_closing || cu->cu_closed) {
555 				error = 0;
556 				cr->cr_error = ESHUTDOWN;
557 			} else {
558 				error = msleep(cr, &cs->cs_lock,
559 				    cu->cu_waitflag, cu->cu_waitchan, tv);
560 			}
561 		} else {
562 			error = EWOULDBLOCK;
563 		}
564 
565 		TAILQ_REMOVE(&cs->cs_pending, cr, cr_link);
566 		cu->cu_sent -= CWNDSCALE;
567 		if (cu->cu_cwnd_wait) {
568 			cu->cu_cwnd_wait = FALSE;
569 			wakeup(&cu->cu_cwnd_wait);
570 		}
571 
572 		if (!error) {
573 			/*
574 			 * We were woken up by the upcall.  If the
575 			 * upcall had a receive error, report that,
576 			 * otherwise we have a reply.
577 			 */
578 			if (cr->cr_error) {
579 				errp->re_errno = cr->cr_error;
580 				errp->re_status = stat = RPC_CANTRECV;
581 				goto out;
582 			}
583 
584 			cu->cu_cwnd += (CWNDSCALE * CWNDSCALE
585 			    + cu->cu_cwnd / 2) / cu->cu_cwnd;
586 			if (cu->cu_cwnd > MAXCWND)
587 				cu->cu_cwnd = MAXCWND;
588 
589 			if (rt) {
590 				/*
591 				 * Add one to the time since a tick
592 				 * count of N means that the actual
593 				 * time taken was somewhere between N
594 				 * and N+1.
595 				 */
596 				rtt = ticks - starttime + 1;
597 
598 				/*
599 				 * Update our estimate of the round
600 				 * trip time using roughly the
601 				 * algorithm described in RFC
602 				 * 2988. Given an RTT sample R:
603 				 *
604 				 * RTTVAR = (1-beta) * RTTVAR + beta * |SRTT-R|
605 				 * SRTT = (1-alpha) * SRTT + alpha * R
606 				 *
607 				 * where alpha = 0.125 and beta = 0.25.
608 				 *
609 				 * The initial retransmit timeout is
610 				 * SRTT + 4*RTTVAR and doubles on each
611 				 * retransmision.
612 				 */
613 				if (rt->rt_srtt == 0) {
614 					rt->rt_srtt = rtt;
615 					rt->rt_deviate = rtt / 2;
616 				} else {
617 					int32_t error = rtt - rt->rt_srtt;
618 					rt->rt_srtt += error / 8;
619 					error = abs(error) - rt->rt_deviate;
620 					rt->rt_deviate += error / 4;
621 				}
622 				rt->rt_rtxcur = rt->rt_srtt + 4*rt->rt_deviate;
623 			}
624 
625 			break;
626 		}
627 
628 		/*
629 		 * The sleep returned an error so our request is still
630 		 * on the list. If we got EWOULDBLOCK, we may want to
631 		 * re-send the request.
632 		 */
633 		if (error != EWOULDBLOCK) {
634 			errp->re_errno = error;
635 			if (error == EINTR || error == ERESTART)
636 				errp->re_status = stat = RPC_INTR;
637 			else
638 				errp->re_status = stat = RPC_CANTRECV;
639 			goto out;
640 		}
641 
642 		time_waited = ticks - starttime;
643 
644 		/* Check for timeout. */
645 		if (time_waited > timeout) {
646 			errp->re_errno = EWOULDBLOCK;
647 			errp->re_status = stat = RPC_TIMEDOUT;
648 			goto out;
649 		}
650 
651 		/* Retransmit if necessary. */
652 		if (time_waited >= next_sendtime) {
653 			cu->cu_cwnd /= 2;
654 			if (cu->cu_cwnd < CWNDSCALE)
655 				cu->cu_cwnd = CWNDSCALE;
656 			if (ext && ext->rc_feedback) {
657 				mtx_unlock(&cs->cs_lock);
658 				if (retrans == 0)
659 					ext->rc_feedback(FEEDBACK_REXMIT1,
660 					    proc, ext->rc_feedback_arg);
661 				else
662 					ext->rc_feedback(FEEDBACK_REXMIT2,
663 					    proc, ext->rc_feedback_arg);
664 				mtx_lock(&cs->cs_lock);
665 			}
666 			if (cu->cu_closing || cu->cu_closed) {
667 				errp->re_errno = ESHUTDOWN;
668 				errp->re_status = stat = RPC_CANTRECV;
669 				goto out;
670 			}
671 			retrans++;
672 			/* update retransmit_time */
673 			if (retransmit_time < RPC_MAX_BACKOFF * hz)
674 				retransmit_time = 2 * retransmit_time;
675 			next_sendtime += retransmit_time;
676 			goto send_again;
677 		}
678 		cu->cu_sent += CWNDSCALE;
679 		TAILQ_INSERT_TAIL(&cs->cs_pending, cr, cr_link);
680 	}
681 
682 got_reply:
683 	/*
684 	 * Now decode and validate the response. We need to drop the
685 	 * lock since xdr_replymsg may end up sleeping in malloc.
686 	 */
687 	mtx_unlock(&cs->cs_lock);
688 
689 	if (ext && ext->rc_feedback)
690 		ext->rc_feedback(FEEDBACK_OK, proc, ext->rc_feedback_arg);
691 
692 	xdrmbuf_create(&xdrs, cr->cr_mrep, XDR_DECODE);
693 	ok = xdr_replymsg(&xdrs, &reply_msg);
694 	cr->cr_mrep = NULL;
695 
696 	if (ok) {
697 		if ((reply_msg.rm_reply.rp_stat == MSG_ACCEPTED) &&
698 		    (reply_msg.acpted_rply.ar_stat == SUCCESS))
699 			errp->re_status = stat = RPC_SUCCESS;
700 		else
701 			stat = _seterr_reply(&reply_msg, &(cu->cu_error));
702 
703 		if (errp->re_status == RPC_SUCCESS) {
704 			results = xdrmbuf_getall(&xdrs);
705 			if (! AUTH_VALIDATE(auth, xid,
706 				&reply_msg.acpted_rply.ar_verf,
707 				&results)) {
708 				errp->re_status = stat = RPC_AUTHERROR;
709 				errp->re_why = AUTH_INVALIDRESP;
710 				if (retrans &&
711 				    auth->ah_cred.oa_flavor == RPCSEC_GSS) {
712 					/*
713 					 * If we retransmitted, its
714 					 * possible that we will
715 					 * receive a reply for one of
716 					 * the earlier transmissions
717 					 * (which will use an older
718 					 * RPCSEC_GSS sequence
719 					 * number). In this case, just
720 					 * go back and listen for a
721 					 * new reply. We could keep a
722 					 * record of all the seq
723 					 * numbers we have transmitted
724 					 * so far so that we could
725 					 * accept a reply for any of
726 					 * them here.
727 					 */
728 					XDR_DESTROY(&xdrs);
729 					mtx_lock(&cs->cs_lock);
730 					cu->cu_sent += CWNDSCALE;
731 					TAILQ_INSERT_TAIL(&cs->cs_pending,
732 					    cr, cr_link);
733 					cr->cr_mrep = NULL;
734 					goto get_reply;
735 				}
736 			} else {
737 				*resultsp = results;
738 			}
739 		}		/* end successful completion */
740 		/*
741 		 * If unsuccessful AND error is an authentication error
742 		 * then refresh credentials and try again, else break
743 		 */
744 		else if (stat == RPC_AUTHERROR)
745 			/* maybe our credentials need to be refreshed ... */
746 			if (nrefreshes > 0 &&
747 			    AUTH_REFRESH(auth, &reply_msg)) {
748 				nrefreshes--;
749 				XDR_DESTROY(&xdrs);
750 				mtx_lock(&cs->cs_lock);
751 				goto call_again;
752 			}
753 		/* end of unsuccessful completion */
754 	}	/* end of valid reply message */
755 	else {
756 		errp->re_status = stat = RPC_CANTDECODERES;
757 
758 	}
759 	XDR_DESTROY(&xdrs);
760 	mtx_lock(&cs->cs_lock);
761 out:
762 	mtx_assert(&cs->cs_lock, MA_OWNED);
763 
764 	if (mreq)
765 		m_freem(mreq);
766 	if (cr->cr_mrep)
767 		m_freem(cr->cr_mrep);
768 
769 	cu->cu_threads--;
770 	if (cu->cu_closing)
771 		wakeup(cu);
772 
773 	mtx_unlock(&cs->cs_lock);
774 
775 	if (auth && stat != RPC_SUCCESS)
776 		AUTH_VALIDATE(auth, xid, NULL, NULL);
777 
778 	free(cr, M_RPC);
779 
780 	return (stat);
781 }
782 
783 static void
784 clnt_dg_geterr(CLIENT *cl, struct rpc_err *errp)
785 {
786 	struct cu_data *cu = (struct cu_data *)cl->cl_private;
787 
788 	*errp = cu->cu_error;
789 }
790 
791 static bool_t
792 clnt_dg_freeres(CLIENT *cl, xdrproc_t xdr_res, void *res_ptr)
793 {
794 	XDR xdrs;
795 	bool_t dummy;
796 
797 	xdrs.x_op = XDR_FREE;
798 	dummy = (*xdr_res)(&xdrs, res_ptr);
799 
800 	return (dummy);
801 }
802 
803 /*ARGSUSED*/
804 static void
805 clnt_dg_abort(CLIENT *h)
806 {
807 }
808 
809 static bool_t
810 clnt_dg_control(CLIENT *cl, u_int request, void *info)
811 {
812 	struct cu_data *cu = (struct cu_data *)cl->cl_private;
813 	struct cu_socket *cs;
814 	struct sockaddr *addr;
815 
816 	cs = cu->cu_socket->so_rcv.sb_upcallarg;
817 	mtx_lock(&cs->cs_lock);
818 
819 	switch (request) {
820 	case CLSET_FD_CLOSE:
821 		cu->cu_closeit = TRUE;
822 		mtx_unlock(&cs->cs_lock);
823 		return (TRUE);
824 	case CLSET_FD_NCLOSE:
825 		cu->cu_closeit = FALSE;
826 		mtx_unlock(&cs->cs_lock);
827 		return (TRUE);
828 	}
829 
830 	/* for other requests which use info */
831 	if (info == NULL) {
832 		mtx_unlock(&cs->cs_lock);
833 		return (FALSE);
834 	}
835 	switch (request) {
836 	case CLSET_TIMEOUT:
837 		if (time_not_ok((struct timeval *)info)) {
838 			mtx_unlock(&cs->cs_lock);
839 			return (FALSE);
840 		}
841 		cu->cu_total = *(struct timeval *)info;
842 		break;
843 	case CLGET_TIMEOUT:
844 		*(struct timeval *)info = cu->cu_total;
845 		break;
846 	case CLSET_RETRY_TIMEOUT:
847 		if (time_not_ok((struct timeval *)info)) {
848 			mtx_unlock(&cs->cs_lock);
849 			return (FALSE);
850 		}
851 		cu->cu_wait = *(struct timeval *)info;
852 		break;
853 	case CLGET_RETRY_TIMEOUT:
854 		*(struct timeval *)info = cu->cu_wait;
855 		break;
856 	case CLGET_SVC_ADDR:
857 		/*
858 		 * Slightly different semantics to userland - we use
859 		 * sockaddr instead of netbuf.
860 		 */
861 		memcpy(info, &cu->cu_raddr, cu->cu_raddr.ss_len);
862 		break;
863 	case CLSET_SVC_ADDR:		/* set to new address */
864 		addr = (struct sockaddr *)info;
865 		(void) memcpy(&cu->cu_raddr, addr, addr->sa_len);
866 		break;
867 	case CLGET_XID:
868 		*(uint32_t *)info = cu->cu_xid;
869 		break;
870 
871 	case CLSET_XID:
872 		/* This will set the xid of the NEXT call */
873 		/* decrement by 1 as clnt_dg_call() increments once */
874 		cu->cu_xid = *(uint32_t *)info - 1;
875 		break;
876 
877 	case CLGET_VERS:
878 		/*
879 		 * This RELIES on the information that, in the call body,
880 		 * the version number field is the fifth field from the
881 		 * beginning of the RPC header. MUST be changed if the
882 		 * call_struct is changed
883 		 */
884 		*(uint32_t *)info =
885 		    ntohl(*(uint32_t *)(void *)(cu->cu_mcallc +
886 		    4 * BYTES_PER_XDR_UNIT));
887 		break;
888 
889 	case CLSET_VERS:
890 		*(uint32_t *)(void *)(cu->cu_mcallc + 4 * BYTES_PER_XDR_UNIT)
891 			= htonl(*(uint32_t *)info);
892 		break;
893 
894 	case CLGET_PROG:
895 		/*
896 		 * This RELIES on the information that, in the call body,
897 		 * the program number field is the fourth field from the
898 		 * beginning of the RPC header. MUST be changed if the
899 		 * call_struct is changed
900 		 */
901 		*(uint32_t *)info =
902 		    ntohl(*(uint32_t *)(void *)(cu->cu_mcallc +
903 		    3 * BYTES_PER_XDR_UNIT));
904 		break;
905 
906 	case CLSET_PROG:
907 		*(uint32_t *)(void *)(cu->cu_mcallc + 3 * BYTES_PER_XDR_UNIT)
908 			= htonl(*(uint32_t *)info);
909 		break;
910 	case CLSET_ASYNC:
911 		cu->cu_async = *(int *)info;
912 		break;
913 	case CLSET_CONNECT:
914 		cu->cu_connect = *(int *)info;
915 		break;
916 	case CLSET_WAITCHAN:
917 		cu->cu_waitchan = (const char *)info;
918 		break;
919 	case CLGET_WAITCHAN:
920 		*(const char **) info = cu->cu_waitchan;
921 		break;
922 	case CLSET_INTERRUPTIBLE:
923 		if (*(int *) info)
924 			cu->cu_waitflag = PCATCH;
925 		else
926 			cu->cu_waitflag = 0;
927 		break;
928 	case CLGET_INTERRUPTIBLE:
929 		if (cu->cu_waitflag)
930 			*(int *) info = TRUE;
931 		else
932 			*(int *) info = FALSE;
933 		break;
934 	default:
935 		mtx_unlock(&cs->cs_lock);
936 		return (FALSE);
937 	}
938 	mtx_unlock(&cs->cs_lock);
939 	return (TRUE);
940 }
941 
942 static void
943 clnt_dg_close(CLIENT *cl)
944 {
945 	struct cu_data *cu = (struct cu_data *)cl->cl_private;
946 	struct cu_socket *cs;
947 	struct cu_request *cr;
948 
949 	cs = cu->cu_socket->so_rcv.sb_upcallarg;
950 	mtx_lock(&cs->cs_lock);
951 
952 	if (cu->cu_closed) {
953 		mtx_unlock(&cs->cs_lock);
954 		return;
955 	}
956 
957 	if (cu->cu_closing) {
958 		while (cu->cu_closing)
959 			msleep(cu, &cs->cs_lock, 0, "rpcclose", 0);
960 		KASSERT(cu->cu_closed, ("client should be closed"));
961 		mtx_unlock(&cs->cs_lock);
962 		return;
963 	}
964 
965 	/*
966 	 * Abort any pending requests and wait until everyone
967 	 * has finished with clnt_vc_call.
968 	 */
969 	cu->cu_closing = TRUE;
970 	TAILQ_FOREACH(cr, &cs->cs_pending, cr_link) {
971 		if (cr->cr_client == cl) {
972 			cr->cr_xid = 0;
973 			cr->cr_error = ESHUTDOWN;
974 			wakeup(cr);
975 		}
976 	}
977 
978 	while (cu->cu_threads)
979 		msleep(cu, &cs->cs_lock, 0, "rpcclose", 0);
980 
981 	cu->cu_closing = FALSE;
982 	cu->cu_closed = TRUE;
983 
984 	mtx_unlock(&cs->cs_lock);
985 	wakeup(cu);
986 }
987 
988 static void
989 clnt_dg_destroy(CLIENT *cl)
990 {
991 	struct cu_data *cu = (struct cu_data *)cl->cl_private;
992 	struct cu_socket *cs;
993 	struct socket *so = NULL;
994 	bool_t lastsocketref;
995 
996 	cs = cu->cu_socket->so_rcv.sb_upcallarg;
997 	clnt_dg_close(cl);
998 
999 	SOCKBUF_LOCK(&cu->cu_socket->so_rcv);
1000 	mtx_lock(&cs->cs_lock);
1001 
1002 	cs->cs_refs--;
1003 	if (cs->cs_refs == 0) {
1004 		mtx_unlock(&cs->cs_lock);
1005 		soupcall_clear(cu->cu_socket, SO_RCV);
1006 		clnt_dg_upcallsdone(cu->cu_socket, cs);
1007 		SOCKBUF_UNLOCK(&cu->cu_socket->so_rcv);
1008 		mtx_destroy(&cs->cs_lock);
1009 		mem_free(cs, sizeof(*cs));
1010 		lastsocketref = TRUE;
1011 	} else {
1012 		mtx_unlock(&cs->cs_lock);
1013 		SOCKBUF_UNLOCK(&cu->cu_socket->so_rcv);
1014 		lastsocketref = FALSE;
1015 	}
1016 
1017 	if (cu->cu_closeit && lastsocketref) {
1018 		so = cu->cu_socket;
1019 		cu->cu_socket = NULL;
1020 	}
1021 
1022 	if (so)
1023 		soclose(so);
1024 
1025 	if (cl->cl_netid && cl->cl_netid[0])
1026 		mem_free(cl->cl_netid, strlen(cl->cl_netid) +1);
1027 	if (cl->cl_tp && cl->cl_tp[0])
1028 		mem_free(cl->cl_tp, strlen(cl->cl_tp) +1);
1029 	mem_free(cu, sizeof (*cu));
1030 	mem_free(cl, sizeof (CLIENT));
1031 }
1032 
1033 /*
1034  * Make sure that the time is not garbage.  -1 value is allowed.
1035  */
1036 static bool_t
1037 time_not_ok(struct timeval *t)
1038 {
1039 	return (t->tv_sec < -1 || t->tv_sec > 100000000 ||
1040 		t->tv_usec < -1 || t->tv_usec > 1000000);
1041 }
1042 
1043 int
1044 clnt_dg_soupcall(struct socket *so, void *arg, int waitflag)
1045 {
1046 	struct cu_socket *cs = (struct cu_socket *) arg;
1047 	struct uio uio;
1048 	struct mbuf *m;
1049 	struct mbuf *control;
1050 	struct cu_request *cr;
1051 	int error, rcvflag, foundreq;
1052 	uint32_t xid;
1053 
1054 	cs->cs_upcallrefs++;
1055 	uio.uio_resid = 1000000000;
1056 	uio.uio_td = curthread;
1057 	do {
1058 		SOCKBUF_UNLOCK(&so->so_rcv);
1059 		m = NULL;
1060 		control = NULL;
1061 		rcvflag = MSG_DONTWAIT;
1062 		error = soreceive(so, NULL, &uio, &m, &control, &rcvflag);
1063 		if (control)
1064 			m_freem(control);
1065 		SOCKBUF_LOCK(&so->so_rcv);
1066 
1067 		if (error == EWOULDBLOCK)
1068 			break;
1069 
1070 		/*
1071 		 * If there was an error, wake up all pending
1072 		 * requests.
1073 		 */
1074 		if (error) {
1075 			mtx_lock(&cs->cs_lock);
1076 			TAILQ_FOREACH(cr, &cs->cs_pending, cr_link) {
1077 				cr->cr_xid = 0;
1078 				cr->cr_error = error;
1079 				wakeup(cr);
1080 			}
1081 			mtx_unlock(&cs->cs_lock);
1082 			break;
1083 		}
1084 
1085 		/*
1086 		 * The XID is in the first uint32_t of the reply.
1087 		 */
1088 		if (m->m_len < sizeof(xid) && m_length(m, NULL) < sizeof(xid)) {
1089 			/*
1090 			 * Should never happen.
1091 			 */
1092 			m_freem(m);
1093 			continue;
1094 		}
1095 
1096 		m_copydata(m, 0, sizeof(xid), (char *)&xid);
1097 		xid = ntohl(xid);
1098 
1099 		/*
1100 		 * Attempt to match this reply with a pending request.
1101 		 */
1102 		mtx_lock(&cs->cs_lock);
1103 		foundreq = 0;
1104 		TAILQ_FOREACH(cr, &cs->cs_pending, cr_link) {
1105 			if (cr->cr_xid == xid) {
1106 				/*
1107 				 * This one matches. We leave the
1108 				 * reply mbuf in cr->cr_mrep. Set the
1109 				 * XID to zero so that we will ignore
1110 				 * any duplicated replies that arrive
1111 				 * before clnt_dg_call removes it from
1112 				 * the queue.
1113 				 */
1114 				cr->cr_xid = 0;
1115 				cr->cr_mrep = m;
1116 				cr->cr_error = 0;
1117 				foundreq = 1;
1118 				wakeup(cr);
1119 				break;
1120 			}
1121 		}
1122 		mtx_unlock(&cs->cs_lock);
1123 
1124 		/*
1125 		 * If we didn't find the matching request, just drop
1126 		 * it - its probably a repeated reply.
1127 		 */
1128 		if (!foundreq)
1129 			m_freem(m);
1130 	} while (m);
1131 	cs->cs_upcallrefs--;
1132 	if (cs->cs_upcallrefs < 0)
1133 		panic("rpcdg upcall refcnt");
1134 	if (cs->cs_upcallrefs == 0)
1135 		wakeup(&cs->cs_upcallrefs);
1136 	return (SU_OK);
1137 }
1138 
1139 /*
1140  * Wait for all upcalls in progress to complete.
1141  */
1142 static void
1143 clnt_dg_upcallsdone(struct socket *so, struct cu_socket *cs)
1144 {
1145 
1146 	SOCKBUF_LOCK_ASSERT(&so->so_rcv);
1147 
1148 	while (cs->cs_upcallrefs > 0)
1149 		(void) msleep(&cs->cs_upcallrefs, SOCKBUF_MTX(&so->so_rcv), 0,
1150 		    "rpcdgup", 0);
1151 }
1152