xref: /openbmc/linux/net/sunrpc/svc.c (revision 6219b20e)
1 // SPDX-License-Identifier: GPL-2.0-only
2 /*
3  * linux/net/sunrpc/svc.c
4  *
5  * High-level RPC service routines
6  *
7  * Copyright (C) 1995, 1996 Olaf Kirch <okir@monad.swb.de>
8  *
9  * Multiple threads pools and NUMAisation
10  * Copyright (c) 2006 Silicon Graphics, Inc.
11  * by Greg Banks <gnb@melbourne.sgi.com>
12  */
13 
14 #include <linux/linkage.h>
15 #include <linux/sched/signal.h>
16 #include <linux/errno.h>
17 #include <linux/net.h>
18 #include <linux/in.h>
19 #include <linux/mm.h>
20 #include <linux/interrupt.h>
21 #include <linux/module.h>
22 #include <linux/kthread.h>
23 #include <linux/slab.h>
24 
25 #include <linux/sunrpc/types.h>
26 #include <linux/sunrpc/xdr.h>
27 #include <linux/sunrpc/stats.h>
28 #include <linux/sunrpc/svcsock.h>
29 #include <linux/sunrpc/clnt.h>
30 #include <linux/sunrpc/bc_xprt.h>
31 
32 #include <trace/events/sunrpc.h>
33 
34 #include "fail.h"
35 
36 #define RPCDBG_FACILITY	RPCDBG_SVCDSP
37 
38 static void svc_unregister(const struct svc_serv *serv, struct net *net);
39 
40 #define svc_serv_is_pooled(serv)    ((serv)->sv_ops->svo_function)
41 
42 #define SVC_POOL_DEFAULT	SVC_POOL_GLOBAL
43 
44 /*
45  * Structure for mapping cpus to pools and vice versa.
46  * Setup once during sunrpc initialisation.
47  */
48 struct svc_pool_map svc_pool_map = {
49 	.mode = SVC_POOL_DEFAULT
50 };
51 EXPORT_SYMBOL_GPL(svc_pool_map);
52 
53 static DEFINE_MUTEX(svc_pool_map_mutex);/* protects svc_pool_map.count only */
54 
55 static int
56 param_set_pool_mode(const char *val, const struct kernel_param *kp)
57 {
58 	int *ip = (int *)kp->arg;
59 	struct svc_pool_map *m = &svc_pool_map;
60 	int err;
61 
62 	mutex_lock(&svc_pool_map_mutex);
63 
64 	err = -EBUSY;
65 	if (m->count)
66 		goto out;
67 
68 	err = 0;
69 	if (!strncmp(val, "auto", 4))
70 		*ip = SVC_POOL_AUTO;
71 	else if (!strncmp(val, "global", 6))
72 		*ip = SVC_POOL_GLOBAL;
73 	else if (!strncmp(val, "percpu", 6))
74 		*ip = SVC_POOL_PERCPU;
75 	else if (!strncmp(val, "pernode", 7))
76 		*ip = SVC_POOL_PERNODE;
77 	else
78 		err = -EINVAL;
79 
80 out:
81 	mutex_unlock(&svc_pool_map_mutex);
82 	return err;
83 }
84 
85 static int
86 param_get_pool_mode(char *buf, const struct kernel_param *kp)
87 {
88 	int *ip = (int *)kp->arg;
89 
90 	switch (*ip)
91 	{
92 	case SVC_POOL_AUTO:
93 		return strlcpy(buf, "auto\n", 20);
94 	case SVC_POOL_GLOBAL:
95 		return strlcpy(buf, "global\n", 20);
96 	case SVC_POOL_PERCPU:
97 		return strlcpy(buf, "percpu\n", 20);
98 	case SVC_POOL_PERNODE:
99 		return strlcpy(buf, "pernode\n", 20);
100 	default:
101 		return sprintf(buf, "%d\n", *ip);
102 	}
103 }
104 
105 module_param_call(pool_mode, param_set_pool_mode, param_get_pool_mode,
106 		 &svc_pool_map.mode, 0644);
107 
108 /*
109  * Detect best pool mapping mode heuristically,
110  * according to the machine's topology.
111  */
112 static int
113 svc_pool_map_choose_mode(void)
114 {
115 	unsigned int node;
116 
117 	if (nr_online_nodes > 1) {
118 		/*
119 		 * Actually have multiple NUMA nodes,
120 		 * so split pools on NUMA node boundaries
121 		 */
122 		return SVC_POOL_PERNODE;
123 	}
124 
125 	node = first_online_node;
126 	if (nr_cpus_node(node) > 2) {
127 		/*
128 		 * Non-trivial SMP, or CONFIG_NUMA on
129 		 * non-NUMA hardware, e.g. with a generic
130 		 * x86_64 kernel on Xeons.  In this case we
131 		 * want to divide the pools on cpu boundaries.
132 		 */
133 		return SVC_POOL_PERCPU;
134 	}
135 
136 	/* default: one global pool */
137 	return SVC_POOL_GLOBAL;
138 }
139 
140 /*
141  * Allocate the to_pool[] and pool_to[] arrays.
142  * Returns 0 on success or an errno.
143  */
144 static int
145 svc_pool_map_alloc_arrays(struct svc_pool_map *m, unsigned int maxpools)
146 {
147 	m->to_pool = kcalloc(maxpools, sizeof(unsigned int), GFP_KERNEL);
148 	if (!m->to_pool)
149 		goto fail;
150 	m->pool_to = kcalloc(maxpools, sizeof(unsigned int), GFP_KERNEL);
151 	if (!m->pool_to)
152 		goto fail_free;
153 
154 	return 0;
155 
156 fail_free:
157 	kfree(m->to_pool);
158 	m->to_pool = NULL;
159 fail:
160 	return -ENOMEM;
161 }
162 
163 /*
164  * Initialise the pool map for SVC_POOL_PERCPU mode.
165  * Returns number of pools or <0 on error.
166  */
167 static int
168 svc_pool_map_init_percpu(struct svc_pool_map *m)
169 {
170 	unsigned int maxpools = nr_cpu_ids;
171 	unsigned int pidx = 0;
172 	unsigned int cpu;
173 	int err;
174 
175 	err = svc_pool_map_alloc_arrays(m, maxpools);
176 	if (err)
177 		return err;
178 
179 	for_each_online_cpu(cpu) {
180 		BUG_ON(pidx >= maxpools);
181 		m->to_pool[cpu] = pidx;
182 		m->pool_to[pidx] = cpu;
183 		pidx++;
184 	}
185 	/* cpus brought online later all get mapped to pool0, sorry */
186 
187 	return pidx;
188 };
189 
190 
191 /*
192  * Initialise the pool map for SVC_POOL_PERNODE mode.
193  * Returns number of pools or <0 on error.
194  */
195 static int
196 svc_pool_map_init_pernode(struct svc_pool_map *m)
197 {
198 	unsigned int maxpools = nr_node_ids;
199 	unsigned int pidx = 0;
200 	unsigned int node;
201 	int err;
202 
203 	err = svc_pool_map_alloc_arrays(m, maxpools);
204 	if (err)
205 		return err;
206 
207 	for_each_node_with_cpus(node) {
208 		/* some architectures (e.g. SN2) have cpuless nodes */
209 		BUG_ON(pidx > maxpools);
210 		m->to_pool[node] = pidx;
211 		m->pool_to[pidx] = node;
212 		pidx++;
213 	}
214 	/* nodes brought online later all get mapped to pool0, sorry */
215 
216 	return pidx;
217 }
218 
219 
220 /*
221  * Add a reference to the global map of cpus to pools (and
222  * vice versa).  Initialise the map if we're the first user.
223  * Returns the number of pools.
224  */
225 unsigned int
226 svc_pool_map_get(void)
227 {
228 	struct svc_pool_map *m = &svc_pool_map;
229 	int npools = -1;
230 
231 	mutex_lock(&svc_pool_map_mutex);
232 
233 	if (m->count++) {
234 		mutex_unlock(&svc_pool_map_mutex);
235 		return m->npools;
236 	}
237 
238 	if (m->mode == SVC_POOL_AUTO)
239 		m->mode = svc_pool_map_choose_mode();
240 
241 	switch (m->mode) {
242 	case SVC_POOL_PERCPU:
243 		npools = svc_pool_map_init_percpu(m);
244 		break;
245 	case SVC_POOL_PERNODE:
246 		npools = svc_pool_map_init_pernode(m);
247 		break;
248 	}
249 
250 	if (npools < 0) {
251 		/* default, or memory allocation failure */
252 		npools = 1;
253 		m->mode = SVC_POOL_GLOBAL;
254 	}
255 	m->npools = npools;
256 
257 	mutex_unlock(&svc_pool_map_mutex);
258 	return m->npools;
259 }
260 EXPORT_SYMBOL_GPL(svc_pool_map_get);
261 
262 /*
263  * Drop a reference to the global map of cpus to pools.
264  * When the last reference is dropped, the map data is
265  * freed; this allows the sysadmin to change the pool
266  * mode using the pool_mode module option without
267  * rebooting or re-loading sunrpc.ko.
268  */
269 void
270 svc_pool_map_put(void)
271 {
272 	struct svc_pool_map *m = &svc_pool_map;
273 
274 	mutex_lock(&svc_pool_map_mutex);
275 
276 	if (!--m->count) {
277 		kfree(m->to_pool);
278 		m->to_pool = NULL;
279 		kfree(m->pool_to);
280 		m->pool_to = NULL;
281 		m->npools = 0;
282 	}
283 
284 	mutex_unlock(&svc_pool_map_mutex);
285 }
286 EXPORT_SYMBOL_GPL(svc_pool_map_put);
287 
288 static int svc_pool_map_get_node(unsigned int pidx)
289 {
290 	const struct svc_pool_map *m = &svc_pool_map;
291 
292 	if (m->count) {
293 		if (m->mode == SVC_POOL_PERCPU)
294 			return cpu_to_node(m->pool_to[pidx]);
295 		if (m->mode == SVC_POOL_PERNODE)
296 			return m->pool_to[pidx];
297 	}
298 	return NUMA_NO_NODE;
299 }
300 /*
301  * Set the given thread's cpus_allowed mask so that it
302  * will only run on cpus in the given pool.
303  */
304 static inline void
305 svc_pool_map_set_cpumask(struct task_struct *task, unsigned int pidx)
306 {
307 	struct svc_pool_map *m = &svc_pool_map;
308 	unsigned int node = m->pool_to[pidx];
309 
310 	/*
311 	 * The caller checks for sv_nrpools > 1, which
312 	 * implies that we've been initialized.
313 	 */
314 	WARN_ON_ONCE(m->count == 0);
315 	if (m->count == 0)
316 		return;
317 
318 	switch (m->mode) {
319 	case SVC_POOL_PERCPU:
320 	{
321 		set_cpus_allowed_ptr(task, cpumask_of(node));
322 		break;
323 	}
324 	case SVC_POOL_PERNODE:
325 	{
326 		set_cpus_allowed_ptr(task, cpumask_of_node(node));
327 		break;
328 	}
329 	}
330 }
331 
332 /*
333  * Use the mapping mode to choose a pool for a given CPU.
334  * Used when enqueueing an incoming RPC.  Always returns
335  * a non-NULL pool pointer.
336  */
337 struct svc_pool *
338 svc_pool_for_cpu(struct svc_serv *serv, int cpu)
339 {
340 	struct svc_pool_map *m = &svc_pool_map;
341 	unsigned int pidx = 0;
342 
343 	/*
344 	 * An uninitialised map happens in a pure client when
345 	 * lockd is brought up, so silently treat it the
346 	 * same as SVC_POOL_GLOBAL.
347 	 */
348 	if (svc_serv_is_pooled(serv)) {
349 		switch (m->mode) {
350 		case SVC_POOL_PERCPU:
351 			pidx = m->to_pool[cpu];
352 			break;
353 		case SVC_POOL_PERNODE:
354 			pidx = m->to_pool[cpu_to_node(cpu)];
355 			break;
356 		}
357 	}
358 	return &serv->sv_pools[pidx % serv->sv_nrpools];
359 }
360 
361 int svc_rpcb_setup(struct svc_serv *serv, struct net *net)
362 {
363 	int err;
364 
365 	err = rpcb_create_local(net);
366 	if (err)
367 		return err;
368 
369 	/* Remove any stale portmap registrations */
370 	svc_unregister(serv, net);
371 	return 0;
372 }
373 EXPORT_SYMBOL_GPL(svc_rpcb_setup);
374 
375 void svc_rpcb_cleanup(struct svc_serv *serv, struct net *net)
376 {
377 	svc_unregister(serv, net);
378 	rpcb_put_local(net);
379 }
380 EXPORT_SYMBOL_GPL(svc_rpcb_cleanup);
381 
382 static int svc_uses_rpcbind(struct svc_serv *serv)
383 {
384 	struct svc_program	*progp;
385 	unsigned int		i;
386 
387 	for (progp = serv->sv_program; progp; progp = progp->pg_next) {
388 		for (i = 0; i < progp->pg_nvers; i++) {
389 			if (progp->pg_vers[i] == NULL)
390 				continue;
391 			if (!progp->pg_vers[i]->vs_hidden)
392 				return 1;
393 		}
394 	}
395 
396 	return 0;
397 }
398 
399 int svc_bind(struct svc_serv *serv, struct net *net)
400 {
401 	if (!svc_uses_rpcbind(serv))
402 		return 0;
403 	return svc_rpcb_setup(serv, net);
404 }
405 EXPORT_SYMBOL_GPL(svc_bind);
406 
407 #if defined(CONFIG_SUNRPC_BACKCHANNEL)
408 static void
409 __svc_init_bc(struct svc_serv *serv)
410 {
411 	INIT_LIST_HEAD(&serv->sv_cb_list);
412 	spin_lock_init(&serv->sv_cb_lock);
413 	init_waitqueue_head(&serv->sv_cb_waitq);
414 }
415 #else
416 static void
417 __svc_init_bc(struct svc_serv *serv)
418 {
419 }
420 #endif
421 
422 /*
423  * Create an RPC service
424  */
425 static struct svc_serv *
426 __svc_create(struct svc_program *prog, unsigned int bufsize, int npools,
427 	     const struct svc_serv_ops *ops)
428 {
429 	struct svc_serv	*serv;
430 	unsigned int vers;
431 	unsigned int xdrsize;
432 	unsigned int i;
433 
434 	if (!(serv = kzalloc(sizeof(*serv), GFP_KERNEL)))
435 		return NULL;
436 	serv->sv_name      = prog->pg_name;
437 	serv->sv_program   = prog;
438 	serv->sv_nrthreads = 1;
439 	serv->sv_stats     = prog->pg_stats;
440 	if (bufsize > RPCSVC_MAXPAYLOAD)
441 		bufsize = RPCSVC_MAXPAYLOAD;
442 	serv->sv_max_payload = bufsize? bufsize : 4096;
443 	serv->sv_max_mesg  = roundup(serv->sv_max_payload + PAGE_SIZE, PAGE_SIZE);
444 	serv->sv_ops = ops;
445 	xdrsize = 0;
446 	while (prog) {
447 		prog->pg_lovers = prog->pg_nvers-1;
448 		for (vers=0; vers<prog->pg_nvers ; vers++)
449 			if (prog->pg_vers[vers]) {
450 				prog->pg_hivers = vers;
451 				if (prog->pg_lovers > vers)
452 					prog->pg_lovers = vers;
453 				if (prog->pg_vers[vers]->vs_xdrsize > xdrsize)
454 					xdrsize = prog->pg_vers[vers]->vs_xdrsize;
455 			}
456 		prog = prog->pg_next;
457 	}
458 	serv->sv_xdrsize   = xdrsize;
459 	INIT_LIST_HEAD(&serv->sv_tempsocks);
460 	INIT_LIST_HEAD(&serv->sv_permsocks);
461 	timer_setup(&serv->sv_temptimer, NULL, 0);
462 	spin_lock_init(&serv->sv_lock);
463 
464 	__svc_init_bc(serv);
465 
466 	serv->sv_nrpools = npools;
467 	serv->sv_pools =
468 		kcalloc(serv->sv_nrpools, sizeof(struct svc_pool),
469 			GFP_KERNEL);
470 	if (!serv->sv_pools) {
471 		kfree(serv);
472 		return NULL;
473 	}
474 
475 	for (i = 0; i < serv->sv_nrpools; i++) {
476 		struct svc_pool *pool = &serv->sv_pools[i];
477 
478 		dprintk("svc: initialising pool %u for %s\n",
479 				i, serv->sv_name);
480 
481 		pool->sp_id = i;
482 		INIT_LIST_HEAD(&pool->sp_sockets);
483 		INIT_LIST_HEAD(&pool->sp_all_threads);
484 		spin_lock_init(&pool->sp_lock);
485 	}
486 
487 	return serv;
488 }
489 
490 struct svc_serv *
491 svc_create(struct svc_program *prog, unsigned int bufsize,
492 	   const struct svc_serv_ops *ops)
493 {
494 	return __svc_create(prog, bufsize, /*npools*/1, ops);
495 }
496 EXPORT_SYMBOL_GPL(svc_create);
497 
498 struct svc_serv *
499 svc_create_pooled(struct svc_program *prog, unsigned int bufsize,
500 		  const struct svc_serv_ops *ops)
501 {
502 	struct svc_serv *serv;
503 	unsigned int npools = svc_pool_map_get();
504 
505 	serv = __svc_create(prog, bufsize, npools, ops);
506 	if (!serv)
507 		goto out_err;
508 	return serv;
509 out_err:
510 	svc_pool_map_put();
511 	return NULL;
512 }
513 EXPORT_SYMBOL_GPL(svc_create_pooled);
514 
515 void svc_shutdown_net(struct svc_serv *serv, struct net *net)
516 {
517 	svc_close_net(serv, net);
518 
519 	if (serv->sv_ops->svo_shutdown)
520 		serv->sv_ops->svo_shutdown(serv, net);
521 }
522 EXPORT_SYMBOL_GPL(svc_shutdown_net);
523 
524 /*
525  * Destroy an RPC service. Should be called with appropriate locking to
526  * protect the sv_nrthreads, sv_permsocks and sv_tempsocks.
527  */
528 void
529 svc_destroy(struct svc_serv *serv)
530 {
531 	dprintk("svc: svc_destroy(%s, %d)\n",
532 				serv->sv_program->pg_name,
533 				serv->sv_nrthreads);
534 
535 	if (serv->sv_nrthreads) {
536 		if (--(serv->sv_nrthreads) != 0) {
537 			svc_sock_update_bufs(serv);
538 			return;
539 		}
540 	} else
541 		printk("svc_destroy: no threads for serv=%p!\n", serv);
542 
543 	del_timer_sync(&serv->sv_temptimer);
544 
545 	/*
546 	 * The last user is gone and thus all sockets have to be destroyed to
547 	 * the point. Check this.
548 	 */
549 	BUG_ON(!list_empty(&serv->sv_permsocks));
550 	BUG_ON(!list_empty(&serv->sv_tempsocks));
551 
552 	cache_clean_deferred(serv);
553 
554 	if (svc_serv_is_pooled(serv))
555 		svc_pool_map_put();
556 
557 	kfree(serv->sv_pools);
558 	kfree(serv);
559 }
560 EXPORT_SYMBOL_GPL(svc_destroy);
561 
562 /*
563  * Allocate an RPC server's buffer space.
564  * We allocate pages and place them in rq_pages.
565  */
566 static int
567 svc_init_buffer(struct svc_rqst *rqstp, unsigned int size, int node)
568 {
569 	unsigned int pages, arghi;
570 
571 	/* bc_xprt uses fore channel allocated buffers */
572 	if (svc_is_backchannel(rqstp))
573 		return 1;
574 
575 	pages = size / PAGE_SIZE + 1; /* extra page as we hold both request and reply.
576 				       * We assume one is at most one page
577 				       */
578 	arghi = 0;
579 	WARN_ON_ONCE(pages > RPCSVC_MAXPAGES);
580 	if (pages > RPCSVC_MAXPAGES)
581 		pages = RPCSVC_MAXPAGES;
582 	while (pages) {
583 		struct page *p = alloc_pages_node(node, GFP_KERNEL, 0);
584 		if (!p)
585 			break;
586 		rqstp->rq_pages[arghi++] = p;
587 		pages--;
588 	}
589 	return pages == 0;
590 }
591 
592 /*
593  * Release an RPC server buffer
594  */
595 static void
596 svc_release_buffer(struct svc_rqst *rqstp)
597 {
598 	unsigned int i;
599 
600 	for (i = 0; i < ARRAY_SIZE(rqstp->rq_pages); i++)
601 		if (rqstp->rq_pages[i])
602 			put_page(rqstp->rq_pages[i]);
603 }
604 
605 struct svc_rqst *
606 svc_rqst_alloc(struct svc_serv *serv, struct svc_pool *pool, int node)
607 {
608 	struct svc_rqst	*rqstp;
609 
610 	rqstp = kzalloc_node(sizeof(*rqstp), GFP_KERNEL, node);
611 	if (!rqstp)
612 		return rqstp;
613 
614 	__set_bit(RQ_BUSY, &rqstp->rq_flags);
615 	spin_lock_init(&rqstp->rq_lock);
616 	rqstp->rq_server = serv;
617 	rqstp->rq_pool = pool;
618 
619 	rqstp->rq_scratch_page = alloc_pages_node(node, GFP_KERNEL, 0);
620 	if (!rqstp->rq_scratch_page)
621 		goto out_enomem;
622 
623 	rqstp->rq_argp = kmalloc_node(serv->sv_xdrsize, GFP_KERNEL, node);
624 	if (!rqstp->rq_argp)
625 		goto out_enomem;
626 
627 	rqstp->rq_resp = kmalloc_node(serv->sv_xdrsize, GFP_KERNEL, node);
628 	if (!rqstp->rq_resp)
629 		goto out_enomem;
630 
631 	if (!svc_init_buffer(rqstp, serv->sv_max_mesg, node))
632 		goto out_enomem;
633 
634 	return rqstp;
635 out_enomem:
636 	svc_rqst_free(rqstp);
637 	return NULL;
638 }
639 EXPORT_SYMBOL_GPL(svc_rqst_alloc);
640 
641 struct svc_rqst *
642 svc_prepare_thread(struct svc_serv *serv, struct svc_pool *pool, int node)
643 {
644 	struct svc_rqst	*rqstp;
645 
646 	rqstp = svc_rqst_alloc(serv, pool, node);
647 	if (!rqstp)
648 		return ERR_PTR(-ENOMEM);
649 
650 	serv->sv_nrthreads++;
651 	spin_lock_bh(&pool->sp_lock);
652 	pool->sp_nrthreads++;
653 	list_add_rcu(&rqstp->rq_all, &pool->sp_all_threads);
654 	spin_unlock_bh(&pool->sp_lock);
655 	return rqstp;
656 }
657 EXPORT_SYMBOL_GPL(svc_prepare_thread);
658 
659 /*
660  * Choose a pool in which to create a new thread, for svc_set_num_threads
661  */
662 static inline struct svc_pool *
663 choose_pool(struct svc_serv *serv, struct svc_pool *pool, unsigned int *state)
664 {
665 	if (pool != NULL)
666 		return pool;
667 
668 	return &serv->sv_pools[(*state)++ % serv->sv_nrpools];
669 }
670 
671 /*
672  * Choose a thread to kill, for svc_set_num_threads
673  */
674 static inline struct task_struct *
675 choose_victim(struct svc_serv *serv, struct svc_pool *pool, unsigned int *state)
676 {
677 	unsigned int i;
678 	struct task_struct *task = NULL;
679 
680 	if (pool != NULL) {
681 		spin_lock_bh(&pool->sp_lock);
682 	} else {
683 		/* choose a pool in round-robin fashion */
684 		for (i = 0; i < serv->sv_nrpools; i++) {
685 			pool = &serv->sv_pools[--(*state) % serv->sv_nrpools];
686 			spin_lock_bh(&pool->sp_lock);
687 			if (!list_empty(&pool->sp_all_threads))
688 				goto found_pool;
689 			spin_unlock_bh(&pool->sp_lock);
690 		}
691 		return NULL;
692 	}
693 
694 found_pool:
695 	if (!list_empty(&pool->sp_all_threads)) {
696 		struct svc_rqst *rqstp;
697 
698 		/*
699 		 * Remove from the pool->sp_all_threads list
700 		 * so we don't try to kill it again.
701 		 */
702 		rqstp = list_entry(pool->sp_all_threads.next, struct svc_rqst, rq_all);
703 		set_bit(RQ_VICTIM, &rqstp->rq_flags);
704 		list_del_rcu(&rqstp->rq_all);
705 		task = rqstp->rq_task;
706 	}
707 	spin_unlock_bh(&pool->sp_lock);
708 
709 	return task;
710 }
711 
712 /* create new threads */
713 static int
714 svc_start_kthreads(struct svc_serv *serv, struct svc_pool *pool, int nrservs)
715 {
716 	struct svc_rqst	*rqstp;
717 	struct task_struct *task;
718 	struct svc_pool *chosen_pool;
719 	unsigned int state = serv->sv_nrthreads-1;
720 	int node;
721 
722 	do {
723 		nrservs--;
724 		chosen_pool = choose_pool(serv, pool, &state);
725 
726 		node = svc_pool_map_get_node(chosen_pool->sp_id);
727 		rqstp = svc_prepare_thread(serv, chosen_pool, node);
728 		if (IS_ERR(rqstp))
729 			return PTR_ERR(rqstp);
730 
731 		__module_get(serv->sv_ops->svo_module);
732 		task = kthread_create_on_node(serv->sv_ops->svo_function, rqstp,
733 					      node, "%s", serv->sv_name);
734 		if (IS_ERR(task)) {
735 			module_put(serv->sv_ops->svo_module);
736 			svc_exit_thread(rqstp);
737 			return PTR_ERR(task);
738 		}
739 
740 		rqstp->rq_task = task;
741 		if (serv->sv_nrpools > 1)
742 			svc_pool_map_set_cpumask(task, chosen_pool->sp_id);
743 
744 		svc_sock_update_bufs(serv);
745 		wake_up_process(task);
746 	} while (nrservs > 0);
747 
748 	return 0;
749 }
750 
751 
752 /* destroy old threads */
753 static int
754 svc_signal_kthreads(struct svc_serv *serv, struct svc_pool *pool, int nrservs)
755 {
756 	struct task_struct *task;
757 	unsigned int state = serv->sv_nrthreads-1;
758 
759 	/* destroy old threads */
760 	do {
761 		task = choose_victim(serv, pool, &state);
762 		if (task == NULL)
763 			break;
764 		send_sig(SIGINT, task, 1);
765 		nrservs++;
766 	} while (nrservs < 0);
767 
768 	return 0;
769 }
770 
771 /*
772  * Create or destroy enough new threads to make the number
773  * of threads the given number.  If `pool' is non-NULL, applies
774  * only to threads in that pool, otherwise round-robins between
775  * all pools.  Caller must ensure that mutual exclusion between this and
776  * server startup or shutdown.
777  *
778  * Destroying threads relies on the service threads filling in
779  * rqstp->rq_task, which only the nfs ones do.  Assumes the serv
780  * has been created using svc_create_pooled().
781  *
782  * Based on code that used to be in nfsd_svc() but tweaked
783  * to be pool-aware.
784  */
785 int
786 svc_set_num_threads(struct svc_serv *serv, struct svc_pool *pool, int nrservs)
787 {
788 	if (pool == NULL) {
789 		/* The -1 assumes caller has done a svc_get() */
790 		nrservs -= (serv->sv_nrthreads-1);
791 	} else {
792 		spin_lock_bh(&pool->sp_lock);
793 		nrservs -= pool->sp_nrthreads;
794 		spin_unlock_bh(&pool->sp_lock);
795 	}
796 
797 	if (nrservs > 0)
798 		return svc_start_kthreads(serv, pool, nrservs);
799 	if (nrservs < 0)
800 		return svc_signal_kthreads(serv, pool, nrservs);
801 	return 0;
802 }
803 EXPORT_SYMBOL_GPL(svc_set_num_threads);
804 
805 /* destroy old threads */
806 static int
807 svc_stop_kthreads(struct svc_serv *serv, struct svc_pool *pool, int nrservs)
808 {
809 	struct task_struct *task;
810 	unsigned int state = serv->sv_nrthreads-1;
811 
812 	/* destroy old threads */
813 	do {
814 		task = choose_victim(serv, pool, &state);
815 		if (task == NULL)
816 			break;
817 		kthread_stop(task);
818 		nrservs++;
819 	} while (nrservs < 0);
820 	return 0;
821 }
822 
823 int
824 svc_set_num_threads_sync(struct svc_serv *serv, struct svc_pool *pool, int nrservs)
825 {
826 	if (pool == NULL) {
827 		/* The -1 assumes caller has done a svc_get() */
828 		nrservs -= (serv->sv_nrthreads-1);
829 	} else {
830 		spin_lock_bh(&pool->sp_lock);
831 		nrservs -= pool->sp_nrthreads;
832 		spin_unlock_bh(&pool->sp_lock);
833 	}
834 
835 	if (nrservs > 0)
836 		return svc_start_kthreads(serv, pool, nrservs);
837 	if (nrservs < 0)
838 		return svc_stop_kthreads(serv, pool, nrservs);
839 	return 0;
840 }
841 EXPORT_SYMBOL_GPL(svc_set_num_threads_sync);
842 
843 /**
844  * svc_rqst_replace_page - Replace one page in rq_pages[]
845  * @rqstp: svc_rqst with pages to replace
846  * @page: replacement page
847  *
848  * When replacing a page in rq_pages, batch the release of the
849  * replaced pages to avoid hammering the page allocator.
850  */
851 void svc_rqst_replace_page(struct svc_rqst *rqstp, struct page *page)
852 {
853 	if (*rqstp->rq_next_page) {
854 		if (!pagevec_space(&rqstp->rq_pvec))
855 			__pagevec_release(&rqstp->rq_pvec);
856 		pagevec_add(&rqstp->rq_pvec, *rqstp->rq_next_page);
857 	}
858 
859 	get_page(page);
860 	*(rqstp->rq_next_page++) = page;
861 }
862 EXPORT_SYMBOL_GPL(svc_rqst_replace_page);
863 
864 /*
865  * Called from a server thread as it's exiting. Caller must hold the "service
866  * mutex" for the service.
867  */
868 void
869 svc_rqst_free(struct svc_rqst *rqstp)
870 {
871 	svc_release_buffer(rqstp);
872 	if (rqstp->rq_scratch_page)
873 		put_page(rqstp->rq_scratch_page);
874 	kfree(rqstp->rq_resp);
875 	kfree(rqstp->rq_argp);
876 	kfree(rqstp->rq_auth_data);
877 	kfree_rcu(rqstp, rq_rcu_head);
878 }
879 EXPORT_SYMBOL_GPL(svc_rqst_free);
880 
881 void
882 svc_exit_thread(struct svc_rqst *rqstp)
883 {
884 	struct svc_serv	*serv = rqstp->rq_server;
885 	struct svc_pool	*pool = rqstp->rq_pool;
886 
887 	spin_lock_bh(&pool->sp_lock);
888 	pool->sp_nrthreads--;
889 	if (!test_and_set_bit(RQ_VICTIM, &rqstp->rq_flags))
890 		list_del_rcu(&rqstp->rq_all);
891 	spin_unlock_bh(&pool->sp_lock);
892 
893 	svc_rqst_free(rqstp);
894 
895 	/* Release the server */
896 	if (serv)
897 		svc_destroy(serv);
898 }
899 EXPORT_SYMBOL_GPL(svc_exit_thread);
900 
901 /*
902  * Register an "inet" protocol family netid with the local
903  * rpcbind daemon via an rpcbind v4 SET request.
904  *
905  * No netconfig infrastructure is available in the kernel, so
906  * we map IP_ protocol numbers to netids by hand.
907  *
908  * Returns zero on success; a negative errno value is returned
909  * if any error occurs.
910  */
911 static int __svc_rpcb_register4(struct net *net, const u32 program,
912 				const u32 version,
913 				const unsigned short protocol,
914 				const unsigned short port)
915 {
916 	const struct sockaddr_in sin = {
917 		.sin_family		= AF_INET,
918 		.sin_addr.s_addr	= htonl(INADDR_ANY),
919 		.sin_port		= htons(port),
920 	};
921 	const char *netid;
922 	int error;
923 
924 	switch (protocol) {
925 	case IPPROTO_UDP:
926 		netid = RPCBIND_NETID_UDP;
927 		break;
928 	case IPPROTO_TCP:
929 		netid = RPCBIND_NETID_TCP;
930 		break;
931 	default:
932 		return -ENOPROTOOPT;
933 	}
934 
935 	error = rpcb_v4_register(net, program, version,
936 					(const struct sockaddr *)&sin, netid);
937 
938 	/*
939 	 * User space didn't support rpcbind v4, so retry this
940 	 * registration request with the legacy rpcbind v2 protocol.
941 	 */
942 	if (error == -EPROTONOSUPPORT)
943 		error = rpcb_register(net, program, version, protocol, port);
944 
945 	return error;
946 }
947 
948 #if IS_ENABLED(CONFIG_IPV6)
949 /*
950  * Register an "inet6" protocol family netid with the local
951  * rpcbind daemon via an rpcbind v4 SET request.
952  *
953  * No netconfig infrastructure is available in the kernel, so
954  * we map IP_ protocol numbers to netids by hand.
955  *
956  * Returns zero on success; a negative errno value is returned
957  * if any error occurs.
958  */
959 static int __svc_rpcb_register6(struct net *net, const u32 program,
960 				const u32 version,
961 				const unsigned short protocol,
962 				const unsigned short port)
963 {
964 	const struct sockaddr_in6 sin6 = {
965 		.sin6_family		= AF_INET6,
966 		.sin6_addr		= IN6ADDR_ANY_INIT,
967 		.sin6_port		= htons(port),
968 	};
969 	const char *netid;
970 	int error;
971 
972 	switch (protocol) {
973 	case IPPROTO_UDP:
974 		netid = RPCBIND_NETID_UDP6;
975 		break;
976 	case IPPROTO_TCP:
977 		netid = RPCBIND_NETID_TCP6;
978 		break;
979 	default:
980 		return -ENOPROTOOPT;
981 	}
982 
983 	error = rpcb_v4_register(net, program, version,
984 					(const struct sockaddr *)&sin6, netid);
985 
986 	/*
987 	 * User space didn't support rpcbind version 4, so we won't
988 	 * use a PF_INET6 listener.
989 	 */
990 	if (error == -EPROTONOSUPPORT)
991 		error = -EAFNOSUPPORT;
992 
993 	return error;
994 }
995 #endif	/* IS_ENABLED(CONFIG_IPV6) */
996 
997 /*
998  * Register a kernel RPC service via rpcbind version 4.
999  *
1000  * Returns zero on success; a negative errno value is returned
1001  * if any error occurs.
1002  */
1003 static int __svc_register(struct net *net, const char *progname,
1004 			  const u32 program, const u32 version,
1005 			  const int family,
1006 			  const unsigned short protocol,
1007 			  const unsigned short port)
1008 {
1009 	int error = -EAFNOSUPPORT;
1010 
1011 	switch (family) {
1012 	case PF_INET:
1013 		error = __svc_rpcb_register4(net, program, version,
1014 						protocol, port);
1015 		break;
1016 #if IS_ENABLED(CONFIG_IPV6)
1017 	case PF_INET6:
1018 		error = __svc_rpcb_register6(net, program, version,
1019 						protocol, port);
1020 #endif
1021 	}
1022 
1023 	trace_svc_register(progname, version, protocol, port, family, error);
1024 	return error;
1025 }
1026 
1027 int svc_rpcbind_set_version(struct net *net,
1028 			    const struct svc_program *progp,
1029 			    u32 version, int family,
1030 			    unsigned short proto,
1031 			    unsigned short port)
1032 {
1033 	return __svc_register(net, progp->pg_name, progp->pg_prog,
1034 				version, family, proto, port);
1035 
1036 }
1037 EXPORT_SYMBOL_GPL(svc_rpcbind_set_version);
1038 
1039 int svc_generic_rpcbind_set(struct net *net,
1040 			    const struct svc_program *progp,
1041 			    u32 version, int family,
1042 			    unsigned short proto,
1043 			    unsigned short port)
1044 {
1045 	const struct svc_version *vers = progp->pg_vers[version];
1046 	int error;
1047 
1048 	if (vers == NULL)
1049 		return 0;
1050 
1051 	if (vers->vs_hidden) {
1052 		trace_svc_noregister(progp->pg_name, version, proto,
1053 				     port, family, 0);
1054 		return 0;
1055 	}
1056 
1057 	/*
1058 	 * Don't register a UDP port if we need congestion
1059 	 * control.
1060 	 */
1061 	if (vers->vs_need_cong_ctrl && proto == IPPROTO_UDP)
1062 		return 0;
1063 
1064 	error = svc_rpcbind_set_version(net, progp, version,
1065 					family, proto, port);
1066 
1067 	return (vers->vs_rpcb_optnl) ? 0 : error;
1068 }
1069 EXPORT_SYMBOL_GPL(svc_generic_rpcbind_set);
1070 
1071 /**
1072  * svc_register - register an RPC service with the local portmapper
1073  * @serv: svc_serv struct for the service to register
1074  * @net: net namespace for the service to register
1075  * @family: protocol family of service's listener socket
1076  * @proto: transport protocol number to advertise
1077  * @port: port to advertise
1078  *
1079  * Service is registered for any address in the passed-in protocol family
1080  */
1081 int svc_register(const struct svc_serv *serv, struct net *net,
1082 		 const int family, const unsigned short proto,
1083 		 const unsigned short port)
1084 {
1085 	struct svc_program	*progp;
1086 	unsigned int		i;
1087 	int			error = 0;
1088 
1089 	WARN_ON_ONCE(proto == 0 && port == 0);
1090 	if (proto == 0 && port == 0)
1091 		return -EINVAL;
1092 
1093 	for (progp = serv->sv_program; progp; progp = progp->pg_next) {
1094 		for (i = 0; i < progp->pg_nvers; i++) {
1095 
1096 			error = progp->pg_rpcbind_set(net, progp, i,
1097 					family, proto, port);
1098 			if (error < 0) {
1099 				printk(KERN_WARNING "svc: failed to register "
1100 					"%sv%u RPC service (errno %d).\n",
1101 					progp->pg_name, i, -error);
1102 				break;
1103 			}
1104 		}
1105 	}
1106 
1107 	return error;
1108 }
1109 
1110 /*
1111  * If user space is running rpcbind, it should take the v4 UNSET
1112  * and clear everything for this [program, version].  If user space
1113  * is running portmap, it will reject the v4 UNSET, but won't have
1114  * any "inet6" entries anyway.  So a PMAP_UNSET should be sufficient
1115  * in this case to clear all existing entries for [program, version].
1116  */
1117 static void __svc_unregister(struct net *net, const u32 program, const u32 version,
1118 			     const char *progname)
1119 {
1120 	int error;
1121 
1122 	error = rpcb_v4_register(net, program, version, NULL, "");
1123 
1124 	/*
1125 	 * User space didn't support rpcbind v4, so retry this
1126 	 * request with the legacy rpcbind v2 protocol.
1127 	 */
1128 	if (error == -EPROTONOSUPPORT)
1129 		error = rpcb_register(net, program, version, 0, 0);
1130 
1131 	trace_svc_unregister(progname, version, error);
1132 }
1133 
1134 /*
1135  * All netids, bind addresses and ports registered for [program, version]
1136  * are removed from the local rpcbind database (if the service is not
1137  * hidden) to make way for a new instance of the service.
1138  *
1139  * The result of unregistration is reported via dprintk for those who want
1140  * verification of the result, but is otherwise not important.
1141  */
1142 static void svc_unregister(const struct svc_serv *serv, struct net *net)
1143 {
1144 	struct svc_program *progp;
1145 	unsigned long flags;
1146 	unsigned int i;
1147 
1148 	clear_thread_flag(TIF_SIGPENDING);
1149 
1150 	for (progp = serv->sv_program; progp; progp = progp->pg_next) {
1151 		for (i = 0; i < progp->pg_nvers; i++) {
1152 			if (progp->pg_vers[i] == NULL)
1153 				continue;
1154 			if (progp->pg_vers[i]->vs_hidden)
1155 				continue;
1156 			__svc_unregister(net, progp->pg_prog, i, progp->pg_name);
1157 		}
1158 	}
1159 
1160 	spin_lock_irqsave(&current->sighand->siglock, flags);
1161 	recalc_sigpending();
1162 	spin_unlock_irqrestore(&current->sighand->siglock, flags);
1163 }
1164 
1165 /*
1166  * dprintk the given error with the address of the client that caused it.
1167  */
1168 #if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
1169 static __printf(2, 3)
1170 void svc_printk(struct svc_rqst *rqstp, const char *fmt, ...)
1171 {
1172 	struct va_format vaf;
1173 	va_list args;
1174 	char 	buf[RPC_MAX_ADDRBUFLEN];
1175 
1176 	va_start(args, fmt);
1177 
1178 	vaf.fmt = fmt;
1179 	vaf.va = &args;
1180 
1181 	dprintk("svc: %s: %pV", svc_print_addr(rqstp, buf, sizeof(buf)), &vaf);
1182 
1183 	va_end(args);
1184 }
1185 #else
1186 static __printf(2,3) void svc_printk(struct svc_rqst *rqstp, const char *fmt, ...) {}
1187 #endif
1188 
1189 static int
1190 svc_generic_dispatch(struct svc_rqst *rqstp, __be32 *statp)
1191 {
1192 	struct kvec *argv = &rqstp->rq_arg.head[0];
1193 	struct kvec *resv = &rqstp->rq_res.head[0];
1194 	const struct svc_procedure *procp = rqstp->rq_procinfo;
1195 
1196 	/*
1197 	 * Decode arguments
1198 	 * XXX: why do we ignore the return value?
1199 	 */
1200 	if (procp->pc_decode &&
1201 	    !procp->pc_decode(rqstp, argv->iov_base)) {
1202 		*statp = rpc_garbage_args;
1203 		return 1;
1204 	}
1205 
1206 	*statp = procp->pc_func(rqstp);
1207 
1208 	if (*statp == rpc_drop_reply ||
1209 	    test_bit(RQ_DROPME, &rqstp->rq_flags))
1210 		return 0;
1211 
1212 	if (rqstp->rq_auth_stat != rpc_auth_ok)
1213 		return 1;
1214 
1215 	if (*statp != rpc_success)
1216 		return 1;
1217 
1218 	/* Encode reply */
1219 	if (procp->pc_encode &&
1220 	    !procp->pc_encode(rqstp, resv->iov_base + resv->iov_len)) {
1221 		dprintk("svc: failed to encode reply\n");
1222 		/* serv->sv_stats->rpcsystemerr++; */
1223 		*statp = rpc_system_err;
1224 	}
1225 	return 1;
1226 }
1227 
1228 __be32
1229 svc_generic_init_request(struct svc_rqst *rqstp,
1230 		const struct svc_program *progp,
1231 		struct svc_process_info *ret)
1232 {
1233 	const struct svc_version *versp = NULL;	/* compiler food */
1234 	const struct svc_procedure *procp = NULL;
1235 
1236 	if (rqstp->rq_vers >= progp->pg_nvers )
1237 		goto err_bad_vers;
1238 	versp = progp->pg_vers[rqstp->rq_vers];
1239 	if (!versp)
1240 		goto err_bad_vers;
1241 
1242 	/*
1243 	 * Some protocol versions (namely NFSv4) require some form of
1244 	 * congestion control.  (See RFC 7530 section 3.1 paragraph 2)
1245 	 * In other words, UDP is not allowed. We mark those when setting
1246 	 * up the svc_xprt, and verify that here.
1247 	 *
1248 	 * The spec is not very clear about what error should be returned
1249 	 * when someone tries to access a server that is listening on UDP
1250 	 * for lower versions. RPC_PROG_MISMATCH seems to be the closest
1251 	 * fit.
1252 	 */
1253 	if (versp->vs_need_cong_ctrl && rqstp->rq_xprt &&
1254 	    !test_bit(XPT_CONG_CTRL, &rqstp->rq_xprt->xpt_flags))
1255 		goto err_bad_vers;
1256 
1257 	if (rqstp->rq_proc >= versp->vs_nproc)
1258 		goto err_bad_proc;
1259 	rqstp->rq_procinfo = procp = &versp->vs_proc[rqstp->rq_proc];
1260 	if (!procp)
1261 		goto err_bad_proc;
1262 
1263 	/* Initialize storage for argp and resp */
1264 	memset(rqstp->rq_argp, 0, procp->pc_argsize);
1265 	memset(rqstp->rq_resp, 0, procp->pc_ressize);
1266 
1267 	/* Bump per-procedure stats counter */
1268 	versp->vs_count[rqstp->rq_proc]++;
1269 
1270 	ret->dispatch = versp->vs_dispatch;
1271 	return rpc_success;
1272 err_bad_vers:
1273 	ret->mismatch.lovers = progp->pg_lovers;
1274 	ret->mismatch.hivers = progp->pg_hivers;
1275 	return rpc_prog_mismatch;
1276 err_bad_proc:
1277 	return rpc_proc_unavail;
1278 }
1279 EXPORT_SYMBOL_GPL(svc_generic_init_request);
1280 
1281 /*
1282  * Common routine for processing the RPC request.
1283  */
1284 static int
1285 svc_process_common(struct svc_rqst *rqstp, struct kvec *argv, struct kvec *resv)
1286 {
1287 	struct svc_program	*progp;
1288 	const struct svc_procedure *procp = NULL;
1289 	struct svc_serv		*serv = rqstp->rq_server;
1290 	struct svc_process_info process;
1291 	__be32			*statp;
1292 	u32			prog, vers;
1293 	__be32			rpc_stat;
1294 	int			auth_res;
1295 	__be32			*reply_statp;
1296 
1297 	rpc_stat = rpc_success;
1298 
1299 	if (argv->iov_len < 6*4)
1300 		goto err_short_len;
1301 
1302 	/* Will be turned off by GSS integrity and privacy services */
1303 	set_bit(RQ_SPLICE_OK, &rqstp->rq_flags);
1304 	/* Will be turned off only when NFSv4 Sessions are used */
1305 	set_bit(RQ_USEDEFERRAL, &rqstp->rq_flags);
1306 	clear_bit(RQ_DROPME, &rqstp->rq_flags);
1307 
1308 	svc_putu32(resv, rqstp->rq_xid);
1309 
1310 	vers = svc_getnl(argv);
1311 
1312 	/* First words of reply: */
1313 	svc_putnl(resv, 1);		/* REPLY */
1314 
1315 	if (vers != 2)		/* RPC version number */
1316 		goto err_bad_rpc;
1317 
1318 	/* Save position in case we later decide to reject: */
1319 	reply_statp = resv->iov_base + resv->iov_len;
1320 
1321 	svc_putnl(resv, 0);		/* ACCEPT */
1322 
1323 	rqstp->rq_prog = prog = svc_getnl(argv);	/* program number */
1324 	rqstp->rq_vers = svc_getnl(argv);	/* version number */
1325 	rqstp->rq_proc = svc_getnl(argv);	/* procedure number */
1326 
1327 	for (progp = serv->sv_program; progp; progp = progp->pg_next)
1328 		if (prog == progp->pg_prog)
1329 			break;
1330 
1331 	/*
1332 	 * Decode auth data, and add verifier to reply buffer.
1333 	 * We do this before anything else in order to get a decent
1334 	 * auth verifier.
1335 	 */
1336 	auth_res = svc_authenticate(rqstp);
1337 	/* Also give the program a chance to reject this call: */
1338 	if (auth_res == SVC_OK && progp)
1339 		auth_res = progp->pg_authenticate(rqstp);
1340 	if (auth_res != SVC_OK)
1341 		trace_svc_authenticate(rqstp, auth_res);
1342 	switch (auth_res) {
1343 	case SVC_OK:
1344 		break;
1345 	case SVC_GARBAGE:
1346 		goto err_garbage;
1347 	case SVC_SYSERR:
1348 		rpc_stat = rpc_system_err;
1349 		goto err_bad;
1350 	case SVC_DENIED:
1351 		goto err_bad_auth;
1352 	case SVC_CLOSE:
1353 		goto close;
1354 	case SVC_DROP:
1355 		goto dropit;
1356 	case SVC_COMPLETE:
1357 		goto sendit;
1358 	}
1359 
1360 	if (progp == NULL)
1361 		goto err_bad_prog;
1362 
1363 	rpc_stat = progp->pg_init_request(rqstp, progp, &process);
1364 	switch (rpc_stat) {
1365 	case rpc_success:
1366 		break;
1367 	case rpc_prog_unavail:
1368 		goto err_bad_prog;
1369 	case rpc_prog_mismatch:
1370 		goto err_bad_vers;
1371 	case rpc_proc_unavail:
1372 		goto err_bad_proc;
1373 	}
1374 
1375 	procp = rqstp->rq_procinfo;
1376 	/* Should this check go into the dispatcher? */
1377 	if (!procp || !procp->pc_func)
1378 		goto err_bad_proc;
1379 
1380 	/* Syntactic check complete */
1381 	serv->sv_stats->rpccnt++;
1382 	trace_svc_process(rqstp, progp->pg_name);
1383 
1384 	/* Build the reply header. */
1385 	statp = resv->iov_base +resv->iov_len;
1386 	svc_putnl(resv, RPC_SUCCESS);
1387 
1388 	/* un-reserve some of the out-queue now that we have a
1389 	 * better idea of reply size
1390 	 */
1391 	if (procp->pc_xdrressize)
1392 		svc_reserve_auth(rqstp, procp->pc_xdrressize<<2);
1393 
1394 	/* Call the function that processes the request. */
1395 	if (!process.dispatch) {
1396 		if (!svc_generic_dispatch(rqstp, statp))
1397 			goto release_dropit;
1398 		if (*statp == rpc_garbage_args)
1399 			goto err_garbage;
1400 	} else {
1401 		dprintk("svc: calling dispatcher\n");
1402 		if (!process.dispatch(rqstp, statp))
1403 			goto release_dropit; /* Release reply info */
1404 	}
1405 
1406 	if (rqstp->rq_auth_stat != rpc_auth_ok)
1407 		goto err_release_bad_auth;
1408 
1409 	/* Check RPC status result */
1410 	if (*statp != rpc_success)
1411 		resv->iov_len = ((void*)statp)  - resv->iov_base + 4;
1412 
1413 	/* Release reply info */
1414 	if (procp->pc_release)
1415 		procp->pc_release(rqstp);
1416 
1417 	if (procp->pc_encode == NULL)
1418 		goto dropit;
1419 
1420  sendit:
1421 	if (svc_authorise(rqstp))
1422 		goto close_xprt;
1423 	return 1;		/* Caller can now send it */
1424 
1425 release_dropit:
1426 	if (procp->pc_release)
1427 		procp->pc_release(rqstp);
1428  dropit:
1429 	svc_authorise(rqstp);	/* doesn't hurt to call this twice */
1430 	dprintk("svc: svc_process dropit\n");
1431 	return 0;
1432 
1433  close:
1434 	svc_authorise(rqstp);
1435 close_xprt:
1436 	if (rqstp->rq_xprt && test_bit(XPT_TEMP, &rqstp->rq_xprt->xpt_flags))
1437 		svc_close_xprt(rqstp->rq_xprt);
1438 	dprintk("svc: svc_process close\n");
1439 	return 0;
1440 
1441 err_short_len:
1442 	svc_printk(rqstp, "short len %zd, dropping request\n",
1443 			argv->iov_len);
1444 	goto close_xprt;
1445 
1446 err_bad_rpc:
1447 	serv->sv_stats->rpcbadfmt++;
1448 	svc_putnl(resv, 1);	/* REJECT */
1449 	svc_putnl(resv, 0);	/* RPC_MISMATCH */
1450 	svc_putnl(resv, 2);	/* Only RPCv2 supported */
1451 	svc_putnl(resv, 2);
1452 	goto sendit;
1453 
1454 err_release_bad_auth:
1455 	if (procp->pc_release)
1456 		procp->pc_release(rqstp);
1457 err_bad_auth:
1458 	dprintk("svc: authentication failed (%d)\n",
1459 		be32_to_cpu(rqstp->rq_auth_stat));
1460 	serv->sv_stats->rpcbadauth++;
1461 	/* Restore write pointer to location of accept status: */
1462 	xdr_ressize_check(rqstp, reply_statp);
1463 	svc_putnl(resv, 1);	/* REJECT */
1464 	svc_putnl(resv, 1);	/* AUTH_ERROR */
1465 	svc_putu32(resv, rqstp->rq_auth_stat);	/* status */
1466 	goto sendit;
1467 
1468 err_bad_prog:
1469 	dprintk("svc: unknown program %d\n", prog);
1470 	serv->sv_stats->rpcbadfmt++;
1471 	svc_putnl(resv, RPC_PROG_UNAVAIL);
1472 	goto sendit;
1473 
1474 err_bad_vers:
1475 	svc_printk(rqstp, "unknown version (%d for prog %d, %s)\n",
1476 		       rqstp->rq_vers, rqstp->rq_prog, progp->pg_name);
1477 
1478 	serv->sv_stats->rpcbadfmt++;
1479 	svc_putnl(resv, RPC_PROG_MISMATCH);
1480 	svc_putnl(resv, process.mismatch.lovers);
1481 	svc_putnl(resv, process.mismatch.hivers);
1482 	goto sendit;
1483 
1484 err_bad_proc:
1485 	svc_printk(rqstp, "unknown procedure (%d)\n", rqstp->rq_proc);
1486 
1487 	serv->sv_stats->rpcbadfmt++;
1488 	svc_putnl(resv, RPC_PROC_UNAVAIL);
1489 	goto sendit;
1490 
1491 err_garbage:
1492 	svc_printk(rqstp, "failed to decode args\n");
1493 
1494 	rpc_stat = rpc_garbage_args;
1495 err_bad:
1496 	serv->sv_stats->rpcbadfmt++;
1497 	svc_putnl(resv, ntohl(rpc_stat));
1498 	goto sendit;
1499 }
1500 
1501 /*
1502  * Process the RPC request.
1503  */
1504 int
1505 svc_process(struct svc_rqst *rqstp)
1506 {
1507 	struct kvec		*argv = &rqstp->rq_arg.head[0];
1508 	struct kvec		*resv = &rqstp->rq_res.head[0];
1509 	struct svc_serv		*serv = rqstp->rq_server;
1510 	u32			dir;
1511 
1512 #if IS_ENABLED(CONFIG_FAIL_SUNRPC)
1513 	if (!fail_sunrpc.ignore_server_disconnect &&
1514 	    should_fail(&fail_sunrpc.attr, 1))
1515 		svc_xprt_deferred_close(rqstp->rq_xprt);
1516 #endif
1517 
1518 	/*
1519 	 * Setup response xdr_buf.
1520 	 * Initially it has just one page
1521 	 */
1522 	rqstp->rq_next_page = &rqstp->rq_respages[1];
1523 	resv->iov_base = page_address(rqstp->rq_respages[0]);
1524 	resv->iov_len = 0;
1525 	rqstp->rq_res.pages = rqstp->rq_respages + 1;
1526 	rqstp->rq_res.len = 0;
1527 	rqstp->rq_res.page_base = 0;
1528 	rqstp->rq_res.page_len = 0;
1529 	rqstp->rq_res.buflen = PAGE_SIZE;
1530 	rqstp->rq_res.tail[0].iov_base = NULL;
1531 	rqstp->rq_res.tail[0].iov_len = 0;
1532 
1533 	dir  = svc_getnl(argv);
1534 	if (dir != 0) {
1535 		/* direction != CALL */
1536 		svc_printk(rqstp, "bad direction %d, dropping request\n", dir);
1537 		serv->sv_stats->rpcbadfmt++;
1538 		goto out_drop;
1539 	}
1540 
1541 	/* Returns 1 for send, 0 for drop */
1542 	if (likely(svc_process_common(rqstp, argv, resv)))
1543 		return svc_send(rqstp);
1544 
1545 out_drop:
1546 	svc_drop(rqstp);
1547 	return 0;
1548 }
1549 EXPORT_SYMBOL_GPL(svc_process);
1550 
1551 #if defined(CONFIG_SUNRPC_BACKCHANNEL)
1552 /*
1553  * Process a backchannel RPC request that arrived over an existing
1554  * outbound connection
1555  */
1556 int
1557 bc_svc_process(struct svc_serv *serv, struct rpc_rqst *req,
1558 	       struct svc_rqst *rqstp)
1559 {
1560 	struct kvec	*argv = &rqstp->rq_arg.head[0];
1561 	struct kvec	*resv = &rqstp->rq_res.head[0];
1562 	struct rpc_task *task;
1563 	int proc_error;
1564 	int error;
1565 
1566 	dprintk("svc: %s(%p)\n", __func__, req);
1567 
1568 	/* Build the svc_rqst used by the common processing routine */
1569 	rqstp->rq_xid = req->rq_xid;
1570 	rqstp->rq_prot = req->rq_xprt->prot;
1571 	rqstp->rq_server = serv;
1572 	rqstp->rq_bc_net = req->rq_xprt->xprt_net;
1573 
1574 	rqstp->rq_addrlen = sizeof(req->rq_xprt->addr);
1575 	memcpy(&rqstp->rq_addr, &req->rq_xprt->addr, rqstp->rq_addrlen);
1576 	memcpy(&rqstp->rq_arg, &req->rq_rcv_buf, sizeof(rqstp->rq_arg));
1577 	memcpy(&rqstp->rq_res, &req->rq_snd_buf, sizeof(rqstp->rq_res));
1578 
1579 	/* Adjust the argument buffer length */
1580 	rqstp->rq_arg.len = req->rq_private_buf.len;
1581 	if (rqstp->rq_arg.len <= rqstp->rq_arg.head[0].iov_len) {
1582 		rqstp->rq_arg.head[0].iov_len = rqstp->rq_arg.len;
1583 		rqstp->rq_arg.page_len = 0;
1584 	} else if (rqstp->rq_arg.len <= rqstp->rq_arg.head[0].iov_len +
1585 			rqstp->rq_arg.page_len)
1586 		rqstp->rq_arg.page_len = rqstp->rq_arg.len -
1587 			rqstp->rq_arg.head[0].iov_len;
1588 	else
1589 		rqstp->rq_arg.len = rqstp->rq_arg.head[0].iov_len +
1590 			rqstp->rq_arg.page_len;
1591 
1592 	/* reset result send buffer "put" position */
1593 	resv->iov_len = 0;
1594 
1595 	/*
1596 	 * Skip the next two words because they've already been
1597 	 * processed in the transport
1598 	 */
1599 	svc_getu32(argv);	/* XID */
1600 	svc_getnl(argv);	/* CALLDIR */
1601 
1602 	/* Parse and execute the bc call */
1603 	proc_error = svc_process_common(rqstp, argv, resv);
1604 
1605 	atomic_dec(&req->rq_xprt->bc_slot_count);
1606 	if (!proc_error) {
1607 		/* Processing error: drop the request */
1608 		xprt_free_bc_request(req);
1609 		error = -EINVAL;
1610 		goto out;
1611 	}
1612 	/* Finally, send the reply synchronously */
1613 	memcpy(&req->rq_snd_buf, &rqstp->rq_res, sizeof(req->rq_snd_buf));
1614 	task = rpc_run_bc_task(req);
1615 	if (IS_ERR(task)) {
1616 		error = PTR_ERR(task);
1617 		goto out;
1618 	}
1619 
1620 	WARN_ON_ONCE(atomic_read(&task->tk_count) != 1);
1621 	error = task->tk_status;
1622 	rpc_put_task(task);
1623 
1624 out:
1625 	dprintk("svc: %s(), error=%d\n", __func__, error);
1626 	return error;
1627 }
1628 EXPORT_SYMBOL_GPL(bc_svc_process);
1629 #endif /* CONFIG_SUNRPC_BACKCHANNEL */
1630 
1631 /*
1632  * Return (transport-specific) limit on the rpc payload.
1633  */
1634 u32 svc_max_payload(const struct svc_rqst *rqstp)
1635 {
1636 	u32 max = rqstp->rq_xprt->xpt_class->xcl_max_payload;
1637 
1638 	if (rqstp->rq_server->sv_max_payload < max)
1639 		max = rqstp->rq_server->sv_max_payload;
1640 	return max;
1641 }
1642 EXPORT_SYMBOL_GPL(svc_max_payload);
1643 
1644 /**
1645  * svc_proc_name - Return RPC procedure name in string form
1646  * @rqstp: svc_rqst to operate on
1647  *
1648  * Return value:
1649  *   Pointer to a NUL-terminated string
1650  */
1651 const char *svc_proc_name(const struct svc_rqst *rqstp)
1652 {
1653 	if (rqstp && rqstp->rq_procinfo)
1654 		return rqstp->rq_procinfo->pc_name;
1655 	return "unknown";
1656 }
1657 
1658 
1659 /**
1660  * svc_encode_result_payload - mark a range of bytes as a result payload
1661  * @rqstp: svc_rqst to operate on
1662  * @offset: payload's byte offset in rqstp->rq_res
1663  * @length: size of payload, in bytes
1664  *
1665  * Returns zero on success, or a negative errno if a permanent
1666  * error occurred.
1667  */
1668 int svc_encode_result_payload(struct svc_rqst *rqstp, unsigned int offset,
1669 			      unsigned int length)
1670 {
1671 	return rqstp->rq_xprt->xpt_ops->xpo_result_payload(rqstp, offset,
1672 							   length);
1673 }
1674 EXPORT_SYMBOL_GPL(svc_encode_result_payload);
1675 
1676 /**
1677  * svc_fill_write_vector - Construct data argument for VFS write call
1678  * @rqstp: svc_rqst to operate on
1679  * @pages: list of pages containing data payload
1680  * @first: buffer containing first section of write payload
1681  * @total: total number of bytes of write payload
1682  *
1683  * Fills in rqstp::rq_vec, and returns the number of elements.
1684  */
1685 unsigned int svc_fill_write_vector(struct svc_rqst *rqstp, struct page **pages,
1686 				   struct kvec *first, size_t total)
1687 {
1688 	struct kvec *vec = rqstp->rq_vec;
1689 	unsigned int i;
1690 
1691 	/* Some types of transport can present the write payload
1692 	 * entirely in rq_arg.pages. In this case, @first is empty.
1693 	 */
1694 	i = 0;
1695 	if (first->iov_len) {
1696 		vec[i].iov_base = first->iov_base;
1697 		vec[i].iov_len = min_t(size_t, total, first->iov_len);
1698 		total -= vec[i].iov_len;
1699 		++i;
1700 	}
1701 
1702 	while (total) {
1703 		vec[i].iov_base = page_address(*pages);
1704 		vec[i].iov_len = min_t(size_t, total, PAGE_SIZE);
1705 		total -= vec[i].iov_len;
1706 		++i;
1707 		++pages;
1708 	}
1709 
1710 	WARN_ON_ONCE(i > ARRAY_SIZE(rqstp->rq_vec));
1711 	return i;
1712 }
1713 EXPORT_SYMBOL_GPL(svc_fill_write_vector);
1714 
1715 /**
1716  * svc_fill_symlink_pathname - Construct pathname argument for VFS symlink call
1717  * @rqstp: svc_rqst to operate on
1718  * @first: buffer containing first section of pathname
1719  * @p: buffer containing remaining section of pathname
1720  * @total: total length of the pathname argument
1721  *
1722  * The VFS symlink API demands a NUL-terminated pathname in mapped memory.
1723  * Returns pointer to a NUL-terminated string, or an ERR_PTR. Caller must free
1724  * the returned string.
1725  */
1726 char *svc_fill_symlink_pathname(struct svc_rqst *rqstp, struct kvec *first,
1727 				void *p, size_t total)
1728 {
1729 	size_t len, remaining;
1730 	char *result, *dst;
1731 
1732 	result = kmalloc(total + 1, GFP_KERNEL);
1733 	if (!result)
1734 		return ERR_PTR(-ESERVERFAULT);
1735 
1736 	dst = result;
1737 	remaining = total;
1738 
1739 	len = min_t(size_t, total, first->iov_len);
1740 	if (len) {
1741 		memcpy(dst, first->iov_base, len);
1742 		dst += len;
1743 		remaining -= len;
1744 	}
1745 
1746 	if (remaining) {
1747 		len = min_t(size_t, remaining, PAGE_SIZE);
1748 		memcpy(dst, p, len);
1749 		dst += len;
1750 	}
1751 
1752 	*dst = '\0';
1753 
1754 	/* Sanity check: Linux doesn't allow the pathname argument to
1755 	 * contain a NUL byte.
1756 	 */
1757 	if (strlen(result) != total) {
1758 		kfree(result);
1759 		return ERR_PTR(-EINVAL);
1760 	}
1761 	return result;
1762 }
1763 EXPORT_SYMBOL_GPL(svc_fill_symlink_pathname);
1764