summaryrefslogtreecommitdiffstats
path: root/sys/nfsserver/nfs_srvkrpc.c
blob: 54b1c4ac5919453b12c8f4e6bf23cc9c5484a790 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
/*-
 * Copyright (c) 1989, 1993
 *	The Regents of the University of California.  All rights reserved.
 *
 * This code is derived from software contributed to Berkeley by
 * Rick Macklem at The University of Guelph.
 *
 * Redistribution and use in source and binary forms, with or without
 * modification, are permitted provided that the following conditions
 * are met:
 * 1. Redistributions of source code must retain the above copyright
 *    notice, this list of conditions and the following disclaimer.
 * 2. Redistributions in binary form must reproduce the above copyright
 *    notice, this list of conditions and the following disclaimer in the
 *    documentation and/or other materials provided with the distribution.
 * 4. Neither the name of the University nor the names of its contributors
 *    may be used to endorse or promote products derived from this software
 *    without specific prior written permission.
 *
 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
 * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
 * SUCH DAMAGE.
 *
 *	@(#)nfs_syscalls.c	8.5 (Berkeley) 3/30/95
 */

#include <sys/cdefs.h>
__FBSDID("$FreeBSD$");

#include "opt_inet6.h"
#include "opt_kgssapi.h"

#include <sys/param.h>
#include <sys/systm.h>
#include <sys/sysproto.h>
#include <sys/kernel.h>
#include <sys/sysctl.h>
#include <sys/file.h>
#include <sys/filedesc.h>
#include <sys/jail.h>
#include <sys/vnode.h>
#include <sys/malloc.h>
#include <sys/mount.h>
#include <sys/priv.h>
#include <sys/proc.h>
#include <sys/bio.h>
#include <sys/buf.h>
#include <sys/mbuf.h>
#include <sys/socket.h>
#include <sys/socketvar.h>
#include <sys/domain.h>
#include <sys/protosw.h>
#include <sys/namei.h>
#include <sys/fcntl.h>
#include <sys/lockf.h>
#include <sys/eventhandler.h>

#include <netinet/in.h>
#include <netinet/tcp.h>
#ifdef INET6
#include <net/if.h>
#include <netinet6/in6_var.h>
#endif

#include <rpc/rpc.h>
#include <rpc/rpcsec_gss.h>
#include <rpc/replay.h>

#include <nfs/xdr_subs.h>
#include <nfs/nfsproto.h>
#include <nfsserver/nfs.h>
#include <nfsserver/nfsm_subs.h>
#include <nfsserver/nfsrvcache.h>
#include <nfsserver/nfs_fha.h>

#include <security/mac/mac_framework.h>

static MALLOC_DEFINE(M_NFSSVC, "nfss_srvsock", "Nfs server structure");

MALLOC_DEFINE(M_NFSRVDESC, "nfss_srvdesc", "NFS server socket descriptor");
MALLOC_DEFINE(M_NFSD, "nfss_daemon", "Nfs server daemon structure");

#define	TRUE	1
#define	FALSE	0

SYSCTL_DECL(_vfs_nfsrv);

SVCPOOL		*nfsrv_pool;
int		nfsd_waiting = 0;
int		nfsrv_numnfsd = 0;
static int	nfs_realign_test;
static int	nfs_realign_count;
struct callout	nfsrv_callout;
static eventhandler_tag nfsrv_nmbclusters_tag;

static int	nfs_privport = 0;
SYSCTL_INT(_vfs_nfsrv, NFS_NFSPRIVPORT, nfs_privport, CTLFLAG_RW,
    &nfs_privport, 0,
    "Only allow clients using a privileged port");
SYSCTL_INT(_vfs_nfsrv, OID_AUTO, gatherdelay, CTLFLAG_RW,
    &nfsrvw_procrastinate, 0,
    "Delay value for write gathering");
SYSCTL_INT(_vfs_nfsrv, OID_AUTO, gatherdelay_v3, CTLFLAG_RW,
    &nfsrvw_procrastinate_v3, 0,
    "Delay in seconds for NFSv3 write gathering");
SYSCTL_INT(_vfs_nfsrv, OID_AUTO, realign_test, CTLFLAG_RW,
	    &nfs_realign_test, 0, "");
SYSCTL_INT(_vfs_nfsrv, OID_AUTO, realign_count, CTLFLAG_RW,
	    &nfs_realign_count, 0, "");

static int	nfssvc_addsock(struct file *, struct thread *);
static int	nfssvc_nfsd(struct thread *, struct nfsd_nfsd_args *);

extern u_long sb_max_adj;

int32_t (*nfsrv3_procs[NFS_NPROCS])(struct nfsrv_descript *nd,
    struct nfssvc_sock *slp, struct mbuf **mreqp) = {
	nfsrv_null,
	nfsrv_getattr,
	nfsrv_setattr,
	nfsrv_lookup,
	nfsrv3_access,
	nfsrv_readlink,
	nfsrv_read,
	nfsrv_write,
	nfsrv_create,
	nfsrv_mkdir,
	nfsrv_symlink,
	nfsrv_mknod,
	nfsrv_remove,
	nfsrv_rmdir,
	nfsrv_rename,
	nfsrv_link,
	nfsrv_readdir,
	nfsrv_readdirplus,
	nfsrv_statfs,
	nfsrv_fsinfo,
	nfsrv_pathconf,
	nfsrv_commit,
	nfsrv_noop
};

/*
 * NFS server system calls
 */
/*
 * This is now called from nfssvc() in nfs/nfs_nfssvc.c.
 */

/*
 * Nfs server psuedo system call for the nfsd's
 * Based on the flag value it either:
 * - adds a socket to the selection list
 * - remains in the kernel as an nfsd
 * - remains in the kernel as an nfsiod
 * For INET6 we suppose that nfsd provides only IN6P_IPV6_V6ONLY sockets
 * and that mountd provides
 *  - sockaddr with no IPv4-mapped addresses
 *  - mask for both INET and INET6 families if there is IPv4-mapped overlap
 */
int
nfssvc_nfsserver(struct thread *td, struct nfssvc_args *uap)
{
	struct file *fp;
	struct nfsd_addsock_args addsockarg;
	struct nfsd_nfsd_args nfsdarg;
	int error;

	if (uap->flag & NFSSVC_ADDSOCK) {
		error = copyin(uap->argp, (caddr_t)&addsockarg,
		    sizeof(addsockarg));
		if (error)
			return (error);
		if ((error = fget(td, addsockarg.sock, &fp)) != 0)
			return (error);
		if (fp->f_type != DTYPE_SOCKET) {
			fdrop(fp, td);
			return (error);	/* XXXRW: Should be EINVAL? */
		}
		error = nfssvc_addsock(fp, td);
		fdrop(fp, td);
	} else if (uap->flag & NFSSVC_OLDNFSD) {
		error = nfssvc_nfsd(td, NULL);
	} else if (uap->flag & NFSSVC_NFSD) {
		if (!uap->argp) 
			return (EINVAL);
		error = copyin(uap->argp, (caddr_t)&nfsdarg,
		    sizeof(nfsdarg));
		if (error)
			return (error);
		error = nfssvc_nfsd(td, &nfsdarg);
	} else {
		error = ENXIO;
	}
	return (error);
}

/*
 * Generate the rpc reply header
 * siz arg. is used to decide if adding a cluster is worthwhile
 */
struct mbuf *
nfs_rephead(int siz, struct nfsrv_descript *nd, int err,
    struct mbuf **mbp, caddr_t *bposp)
{
	u_int32_t *tl;
	struct mbuf *mreq;
	caddr_t bpos;
	struct mbuf *mb;

	if (err == EBADRPC)
		return (NULL);

	nd->nd_repstat = err;
	if (err && (nd->nd_flag & ND_NFSV3) == 0)	/* XXX recheck */
		siz = 0;

	MGET(mreq, M_WAIT, MT_DATA);

	/*
	 * If this is a big reply, use a cluster
	 */
	mreq->m_len = 0;
	if (siz >= MINCLSIZE) {
		MCLGET(mreq, M_WAIT);
	}
	mb = mreq;
	bpos = mtod(mb, caddr_t);

	if (err != NFSERR_RETVOID) {
		tl = nfsm_build(u_int32_t *, NFSX_UNSIGNED);
		if (err)
			*tl = txdr_unsigned(nfsrv_errmap(nd, err));
		else
			*tl = 0;
	}

	*mbp = mb;
	*bposp = bpos;
	if (err != 0 && err != NFSERR_RETVOID)
		nfsrvstats.srvrpc_errs++;

	return (mreq);
}

/*
 *	nfs_realign:
 *
 *	Check for badly aligned mbuf data and realign by copying the unaligned
 *	portion of the data into a new mbuf chain and freeing the portions
 *	of the old chain that were replaced.
 *
 *	We cannot simply realign the data within the existing mbuf chain
 *	because the underlying buffers may contain other rpc commands and
 *	we cannot afford to overwrite them.
 *
 *	We would prefer to avoid this situation entirely.  The situation does
 *	not occur with NFS/UDP and is supposed to only occassionally occur
 *	with TCP.  Use vfs.nfs.realign_count and realign_test to check this.
 */
static void
nfs_realign(struct mbuf **pm)	/* XXX COMMON */
{
	struct mbuf *m;
	struct mbuf *n = NULL;
	int off = 0;

	++nfs_realign_test;
	while ((m = *pm) != NULL) {
		if ((m->m_len & 0x3) || (mtod(m, intptr_t) & 0x3)) {
			MGET(n, M_WAIT, MT_DATA);
			if (m->m_len >= MINCLSIZE) {
				MCLGET(n, M_WAIT);
			}
			n->m_len = 0;
			break;
		}
		pm = &m->m_next;
	}

	/*
	 * If n is non-NULL, loop on m copying data, then replace the
	 * portion of the chain that had to be realigned.
	 */
	if (n != NULL) {
		++nfs_realign_count;
		while (m) {
			m_copyback(n, off, m->m_len, mtod(m, caddr_t));
			off += m->m_len;
			m = m->m_next;
		}
		m_freem(*pm);
		*pm = n;
	}
}

static void
nfssvc_program(struct svc_req *rqst, SVCXPRT *xprt)
{
	rpcproc_t procnum;
	int32_t (*proc)(struct nfsrv_descript *nd, struct nfssvc_sock *slp,
	    struct mbuf **mreqp);
	int flag;
	struct nfsrv_descript nd;
	struct mbuf *mreq, *mrep;
	int error;

	if (rqst->rq_vers == NFS_VER2) {
		if (rqst->rq_proc > NFSV2PROC_STATFS) {
			svcerr_noproc(rqst);
			svc_freereq(rqst);
			return;
		}
		procnum = nfsrv_nfsv3_procid[rqst->rq_proc];
		flag = 0;
	} else {
		if (rqst->rq_proc >= NFS_NPROCS) {
			svcerr_noproc(rqst);
			svc_freereq(rqst);
			return;
		}
		procnum = rqst->rq_proc;
		flag = ND_NFSV3;
	}
	proc = nfsrv3_procs[procnum];

	mreq = mrep = NULL;
	mreq = rqst->rq_args;
	rqst->rq_args = NULL;
	nfs_realign(&mreq);

	/*
	 * Note: we want rq_addr, not svc_getrpccaller for nd_nam2 -
	 * NFS_SRVMAXDATA uses a NULL value for nd_nam2 to detect TCP
	 * mounts.
	 */
	memset(&nd, 0, sizeof(nd));
	nd.nd_md = nd.nd_mrep = mreq;
	nd.nd_dpos = mtod(mreq, caddr_t);
	nd.nd_nam = svc_getrpccaller(rqst);
	nd.nd_nam2 = rqst->rq_addr;
	nd.nd_procnum = procnum;
	nd.nd_cr = NULL;
	nd.nd_flag = flag;

	if (nfs_privport) {
		/* Check if source port is privileged */
		u_short port;
		struct sockaddr *nam = nd.nd_nam;
		struct sockaddr_in *sin;

		sin = (struct sockaddr_in *)nam;
		/*
		 * INET/INET6 - same code:
		 *    sin_port and sin6_port are at same offset
		 */
		port = ntohs(sin->sin_port);
		if (port >= IPPORT_RESERVED &&
		    nd.nd_procnum != NFSPROC_NULL) {
#ifdef INET6
			char b6[INET6_ADDRSTRLEN];
#if defined(KLD_MODULE)
			/* Do not use ip6_sprintf: the nfs module should work without INET6. */
#define ip6_sprintf(buf, a)						\
			(sprintf((buf), "%x:%x:%x:%x:%x:%x:%x:%x",	\
			    (a)->s6_addr16[0], (a)->s6_addr16[1],	\
			    (a)->s6_addr16[2], (a)->s6_addr16[3],	\
			    (a)->s6_addr16[4], (a)->s6_addr16[5],	\
			    (a)->s6_addr16[6], (a)->s6_addr16[7]),	\
			    (buf))
#endif
#endif
			printf("NFS request from unprivileged port (%s:%d)\n",
#ifdef INET6
			    sin->sin_family == AF_INET6 ?
			    ip6_sprintf(b6, &satosin6(sin)->sin6_addr) :
#if defined(KLD_MODULE)
#undef ip6_sprintf
#endif
#endif
			    inet_ntoa(sin->sin_addr), port);
			m_freem(mreq);
			svcerr_weakauth(rqst);
			svc_freereq(rqst);
			return;
		}
	}

	if (proc != nfsrv_null) {
		if (!svc_getcred(rqst, &nd.nd_cr, &nd.nd_credflavor)) {
			m_freem(mreq);
			svcerr_weakauth(rqst);
			svc_freereq(rqst);
			return;
		}
#ifdef MAC
		mac_cred_associate_nfsd(nd.nd_cr);
#endif
	}
	nfsrvstats.srvrpccnt[nd.nd_procnum]++;

	error = proc(&nd, NULL, &mrep);

	if (nd.nd_cr)
		crfree(nd.nd_cr);

	if (mrep == NULL) {
		svcerr_decode(rqst);
		svc_freereq(rqst);
		return;
	}
	if (error && error != NFSERR_RETVOID) {
		svcerr_systemerr(rqst);
		svc_freereq(rqst);
		return;
	}
	if (nd.nd_repstat & NFSERR_AUTHERR) {
		svcerr_auth(rqst, nd.nd_repstat & ~NFSERR_AUTHERR);
		m_freem(mrep);
	} else {
		if (!svc_sendreply_mbuf(rqst, mrep))
			svcerr_systemerr(rqst);
	}
	svc_freereq(rqst);
}

/*
 * Adds a socket to the list for servicing by nfsds.
 */
static int
nfssvc_addsock(struct file *fp, struct thread *td)
{
	int siz;
	struct socket *so;
	int error;
	SVCXPRT *xprt;

	so = fp->f_data;

	siz = sb_max_adj;
	error = soreserve(so, siz, siz);
	if (error) {
		return (error);
	}

	/*
	 * Steal the socket from userland so that it doesn't close
	 * unexpectedly.
	 */
	if (so->so_type == SOCK_DGRAM)
		xprt = svc_dg_create(nfsrv_pool, so, 0, 0);
	else
		xprt = svc_vc_create(nfsrv_pool, so, 0, 0);
	if (xprt) {
		fp->f_ops = &badfileops;
		fp->f_data = NULL;
		svc_reg(xprt, NFS_PROG, NFS_VER2, nfssvc_program, NULL);
		svc_reg(xprt, NFS_PROG, NFS_VER3, nfssvc_program, NULL);
		SVC_RELEASE(xprt);
	}

	return (0);
}

/*
 * Called by nfssvc() for nfsds. Just loops around servicing rpc requests
 * until it is killed by a signal.
 */
static int
nfssvc_nfsd(struct thread *td, struct nfsd_nfsd_args *args)
{
#ifdef KGSSAPI
	char principal[128];
	int error;
#endif

#ifdef KGSSAPI
	if (args) {
		error = copyinstr(args->principal, principal,
		    sizeof(principal), NULL);
		if (error)
			return (error);
	} else {
		memcpy(principal, "nfs@", 4);
		getcredhostname(td->td_ucred, principal + 4,
		    sizeof(principal) - 4);
	}
#endif

	/*
	 * Only the first nfsd actually does any work. The RPC code
	 * adds threads to it as needed. Any extra processes offered
	 * by nfsd just exit. If nfsd is new enough, it will call us
	 * once with a structure that specifies how many threads to
	 * use.
	 */
	NFSD_LOCK();
	if (nfsrv_numnfsd == 0) {
		nfsrv_numnfsd++;

		NFSD_UNLOCK();

#ifdef KGSSAPI
		rpc_gss_set_svc_name(principal, "kerberosv5",
		    GSS_C_INDEFINITE, NFS_PROG, NFS_VER2);
		rpc_gss_set_svc_name(principal, "kerberosv5",
		    GSS_C_INDEFINITE, NFS_PROG, NFS_VER3);
#endif

		if (args) {
			nfsrv_pool->sp_minthreads = args->minthreads;
			nfsrv_pool->sp_maxthreads = args->maxthreads;
		} else {
			nfsrv_pool->sp_minthreads = 4;
			nfsrv_pool->sp_maxthreads = 4;
		}
			
		svc_run(nfsrv_pool);

#ifdef KGSSAPI
		rpc_gss_clear_svc_name(NFS_PROG, NFS_VER2);
		rpc_gss_clear_svc_name(NFS_PROG, NFS_VER3);
#endif

		NFSD_LOCK();
		nfsrv_numnfsd--;
		nfsrv_init(TRUE);
	}
	NFSD_UNLOCK();

	return (0);
}

/*
 * Size the NFS server's duplicate request cache at 1/2 the
 * nmbclusters, floating within a (64, 2048) range. This is to
 * prevent all mbuf clusters being tied up in the NFS dupreq
 * cache for small values of nmbclusters.
 */
static size_t
nfsrv_replay_size(void)
{
	size_t replaysiz;

	replaysiz = nmbclusters / 2;
	if (replaysiz > NFSRVCACHE_MAX_SIZE)
		replaysiz = NFSRVCACHE_MAX_SIZE;
	if (replaysiz < NFSRVCACHE_MIN_SIZE)
		replaysiz = NFSRVCACHE_MIN_SIZE;
	replaysiz *= MCLBYTES;

	return (replaysiz);
}

/*
 * Called when nmbclusters changes - we resize the replay cache
 * accordingly.
 */
static void
nfsrv_nmbclusters_change(void *tag)
{

	if (nfsrv_pool)
		replay_setsize(nfsrv_pool->sp_rcache, nfsrv_replay_size());
}

/*
 * Initialize the data structures for the server.
 * Handshake with any new nfsds starting up to avoid any chance of
 * corruption.
 */
void
nfsrv_init(int terminating)
{

	NFSD_LOCK_ASSERT();

	if (terminating) {
		NFSD_UNLOCK();
		EVENTHANDLER_DEREGISTER(nmbclusters_change,
		    nfsrv_nmbclusters_tag);
		svcpool_destroy(nfsrv_pool);
		nfsrv_pool = NULL;
		NFSD_LOCK();
	} else
		nfs_pub.np_valid = 0;

	NFSD_UNLOCK();

	nfsrv_pool = svcpool_create("nfsd", SYSCTL_STATIC_CHILDREN(_vfs_nfsrv));
	nfsrv_pool->sp_rcache = replay_newcache(nfsrv_replay_size());
	nfsrv_pool->sp_assign = fha_assign;
	nfsrv_pool->sp_done = fha_nd_complete;
	nfsrv_nmbclusters_tag = EVENTHANDLER_REGISTER(nmbclusters_change,
	    nfsrv_nmbclusters_change, NULL, EVENTHANDLER_PRI_FIRST);

	NFSD_LOCK();
}
OpenPOWER on IntegriCloud