summaryrefslogtreecommitdiffstats
path: root/sys/nfsclient/nfs_subs.c
blob: 361375684cf07280c92b555adeab9aacf34074be (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
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
/*-
 * 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_subs.c  8.8 (Berkeley) 5/22/95
 */

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

/*
 * These functions support the macros and help fiddle mbuf chains for
 * the nfs op functions. They do things like create the rpc header and
 * copy data between mbuf chains and uio lists.
 */

#include "opt_kdtrace.h"

#include <sys/param.h>
#include <sys/systm.h>
#include <sys/kernel.h>
#include <sys/bio.h>
#include <sys/buf.h>
#include <sys/proc.h>
#include <sys/mount.h>
#include <sys/vnode.h>
#include <sys/namei.h>
#include <sys/mbuf.h>
#include <sys/socket.h>
#include <sys/stat.h>
#include <sys/malloc.h>
#include <sys/sysent.h>
#include <sys/syscall.h>
#include <sys/sysproto.h>

#include <vm/vm.h>
#include <vm/vm_object.h>
#include <vm/vm_extern.h>
#include <vm/uma.h>

#include <nfs/nfsproto.h>
#include <nfsclient/nfs.h>
#include <nfsclient/nfsnode.h>
#include <nfsclient/nfs_kdtrace.h>
#include <nfs/xdr_subs.h>
#include <nfsclient/nfsm_subs.h>
#include <nfsclient/nfsmount.h>

#include <netinet/in.h>

/*
 * Note that stdarg.h and the ANSI style va_start macro is used for both
 * ANSI and traditional C compilers.
 */
#include <machine/stdarg.h>

#ifdef KDTRACE_HOOKS
dtrace_nfsclient_attrcache_flush_probe_func_t
    dtrace_nfsclient_attrcache_flush_done_probe;
uint32_t nfsclient_attrcache_flush_done_id;

dtrace_nfsclient_attrcache_get_hit_probe_func_t
    dtrace_nfsclient_attrcache_get_hit_probe;
uint32_t nfsclient_attrcache_get_hit_id;

dtrace_nfsclient_attrcache_get_miss_probe_func_t
    dtrace_nfsclient_attrcache_get_miss_probe;
uint32_t nfsclient_attrcache_get_miss_id;

dtrace_nfsclient_attrcache_load_probe_func_t
    dtrace_nfsclient_attrcache_load_done_probe;
uint32_t nfsclient_attrcache_load_done_id;
#endif /* !KDTRACE_HOOKS */

/*
 * Data items converted to xdr at startup, since they are constant
 * This is kinda hokey, but may save a little time doing byte swaps
 */
u_int32_t	nfs_xdrneg1;
u_int32_t	nfs_true, nfs_false;

/* And other global data */
static u_int32_t nfs_xid = 0;
static enum vtype nv2tov_type[8]= {
	VNON, VREG, VDIR, VBLK, VCHR, VLNK, VNON,  VNON
};

int		nfs_ticks;
int		nfs_pbuf_freecnt = -1;	/* start out unlimited */

struct nfs_bufq	nfs_bufq;
static struct mtx nfs_xid_mtx;

/*
 * and the reverse mapping from generic to Version 2 procedure numbers
 */
int nfsv2_procid[NFS_NPROCS] = {
	NFSV2PROC_NULL,
	NFSV2PROC_GETATTR,
	NFSV2PROC_SETATTR,
	NFSV2PROC_LOOKUP,
	NFSV2PROC_NOOP,
	NFSV2PROC_READLINK,
	NFSV2PROC_READ,
	NFSV2PROC_WRITE,
	NFSV2PROC_CREATE,
	NFSV2PROC_MKDIR,
	NFSV2PROC_SYMLINK,
	NFSV2PROC_CREATE,
	NFSV2PROC_REMOVE,
	NFSV2PROC_RMDIR,
	NFSV2PROC_RENAME,
	NFSV2PROC_LINK,
	NFSV2PROC_READDIR,
	NFSV2PROC_NOOP,
	NFSV2PROC_STATFS,
	NFSV2PROC_NOOP,
	NFSV2PROC_NOOP,
	NFSV2PROC_NOOP,
	NFSV2PROC_NOOP,
};

LIST_HEAD(nfsnodehashhead, nfsnode);

u_int32_t
nfs_xid_gen(void)
{
	uint32_t xid;

	mtx_lock(&nfs_xid_mtx);

	/* Get a pretty random xid to start with */
	if (!nfs_xid)
		nfs_xid = random();
	/*
	 * Skip zero xid if it should ever happen.
	 */
	if (++nfs_xid == 0)
		nfs_xid++;
	xid = nfs_xid;
	mtx_unlock(&nfs_xid_mtx);
	return xid;
}

/*
 * Create the header for an rpc request packet
 * The hsiz is the size of the rest of the nfs request header.
 * (just used to decide if a cluster is a good idea)
 */
struct mbuf *
nfsm_reqhead(struct vnode *vp, u_long procid, int hsiz)
{
	struct mbuf *mb;

	MGET(mb, M_WAIT, MT_DATA);
	if (hsiz >= MINCLSIZE)
		MCLGET(mb, M_WAIT);
	mb->m_len = 0;
	return (mb);
}

/*
 * copies a uio scatter/gather list to an mbuf chain.
 * NOTE: can ony handle iovcnt == 1
 */
int
nfsm_uiotombuf(struct uio *uiop, struct mbuf **mq, int siz, caddr_t *bpos)
{
	char *uiocp;
	struct mbuf *mp, *mp2;
	int xfer, left, mlen;
	int uiosiz, clflg, rem;
	char *cp;

	KASSERT(uiop->uio_iovcnt == 1, ("nfsm_uiotombuf: iovcnt != 1"));

	if (siz > MLEN)		/* or should it >= MCLBYTES ?? */
		clflg = 1;
	else
		clflg = 0;
	rem = nfsm_rndup(siz)-siz;
	mp = mp2 = *mq;
	while (siz > 0) {
		left = uiop->uio_iov->iov_len;
		uiocp = uiop->uio_iov->iov_base;
		if (left > siz)
			left = siz;
		uiosiz = left;
		while (left > 0) {
			mlen = M_TRAILINGSPACE(mp);
			if (mlen == 0) {
				MGET(mp, M_WAIT, MT_DATA);
				if (clflg)
					MCLGET(mp, M_WAIT);
				mp->m_len = 0;
				mp2->m_next = mp;
				mp2 = mp;
				mlen = M_TRAILINGSPACE(mp);
			}
			xfer = (left > mlen) ? mlen : left;
#ifdef notdef
			/* Not Yet.. */
			if (uiop->uio_iov->iov_op != NULL)
				(*(uiop->uio_iov->iov_op))
				(uiocp, mtod(mp, caddr_t)+mp->m_len, xfer);
			else
#endif
			if (uiop->uio_segflg == UIO_SYSSPACE)
				bcopy(uiocp, mtod(mp, caddr_t)+mp->m_len, xfer);
			else
				copyin(uiocp, mtod(mp, caddr_t)+mp->m_len, xfer);
			mp->m_len += xfer;
			left -= xfer;
			uiocp += xfer;
			uiop->uio_offset += xfer;
			uiop->uio_resid -= xfer;
		}
		uiop->uio_iov->iov_base =
		    (char *)uiop->uio_iov->iov_base + uiosiz;
		uiop->uio_iov->iov_len -= uiosiz;
		siz -= uiosiz;
	}
	if (rem > 0) {
		if (rem > M_TRAILINGSPACE(mp)) {
			MGET(mp, M_WAIT, MT_DATA);
			mp->m_len = 0;
			mp2->m_next = mp;
		}
		cp = mtod(mp, caddr_t)+mp->m_len;
		for (left = 0; left < rem; left++)
			*cp++ = '\0';
		mp->m_len += rem;
		*bpos = cp;
	} else
		*bpos = mtod(mp, caddr_t)+mp->m_len;
	*mq = mp;
	return (0);
}

/*
 * Copy a string into mbufs for the hard cases...
 */
int
nfsm_strtmbuf(struct mbuf **mb, char **bpos, const char *cp, long siz)
{
	struct mbuf *m1 = NULL, *m2;
	long left, xfer, len, tlen;
	u_int32_t *tl;
	int putsize;

	putsize = 1;
	m2 = *mb;
	left = M_TRAILINGSPACE(m2);
	if (left > 0) {
		tl = ((u_int32_t *)(*bpos));
		*tl++ = txdr_unsigned(siz);
		putsize = 0;
		left -= NFSX_UNSIGNED;
		m2->m_len += NFSX_UNSIGNED;
		if (left > 0) {
			bcopy(cp, (caddr_t) tl, left);
			siz -= left;
			cp += left;
			m2->m_len += left;
			left = 0;
		}
	}
	/* Loop around adding mbufs */
	while (siz > 0) {
		MGET(m1, M_WAIT, MT_DATA);
		if (siz > MLEN)
			MCLGET(m1, M_WAIT);
		m1->m_len = NFSMSIZ(m1);
		m2->m_next = m1;
		m2 = m1;
		tl = mtod(m1, u_int32_t *);
		tlen = 0;
		if (putsize) {
			*tl++ = txdr_unsigned(siz);
			m1->m_len -= NFSX_UNSIGNED;
			tlen = NFSX_UNSIGNED;
			putsize = 0;
		}
		if (siz < m1->m_len) {
			len = nfsm_rndup(siz);
			xfer = siz;
			if (xfer < len)
				*(tl+(xfer>>2)) = 0;
		} else {
			xfer = len = m1->m_len;
		}
		bcopy(cp, (caddr_t) tl, xfer);
		m1->m_len = len+tlen;
		siz -= xfer;
		cp += xfer;
	}
	*mb = m1;
	*bpos = mtod(m1, caddr_t)+m1->m_len;
	return (0);
}

/*
 * Called once to initialize data structures...
 */
int
nfs_init(struct vfsconf *vfsp)
{
	int i;

	nfsmount_zone = uma_zcreate("NFSMOUNT", sizeof(struct nfsmount),
	    NULL, NULL, NULL, NULL, UMA_ALIGN_PTR, 0);
	nfs_true = txdr_unsigned(TRUE);
	nfs_false = txdr_unsigned(FALSE);
	nfs_xdrneg1 = txdr_unsigned(-1);
	nfs_ticks = (hz * NFS_TICKINTVL + 500) / 1000;
	if (nfs_ticks < 1)
		nfs_ticks = 1;
	/* Ensure async daemons disabled */
	for (i = 0; i < NFS_MAXASYNCDAEMON; i++) {
		nfs_iodwant[i] = NFSIOD_NOT_AVAILABLE;
		nfs_iodmount[i] = NULL;
	}
	nfs_nhinit();			/* Init the nfsnode table */

	/*
	 * Initialize reply list and start timer
	 */
	mtx_init(&nfs_iod_mtx, "NFS iod lock", NULL, MTX_DEF);
	mtx_init(&nfs_xid_mtx, "NFS xid lock", NULL, MTX_DEF);

	nfs_pbuf_freecnt = nswbuf / 2 + 1;

	return (0);
}

int
nfs_uninit(struct vfsconf *vfsp)
{
	int i;

	/*
	 * Tell all nfsiod processes to exit. Clear nfs_iodmax, and wakeup
	 * any sleeping nfsiods so they check nfs_iodmax and exit.
	 */
	mtx_lock(&nfs_iod_mtx);
	nfs_iodmax = 0;
	for (i = 0; i < nfs_numasync; i++)
		if (nfs_iodwant[i] == NFSIOD_AVAILABLE)
			wakeup(&nfs_iodwant[i]);
	/* The last nfsiod to exit will wake us up when nfs_numasync hits 0 */
	while (nfs_numasync)
		msleep(&nfs_numasync, &nfs_iod_mtx, PWAIT, "ioddie", 0);
	mtx_unlock(&nfs_iod_mtx);
	nfs_nhuninit();
	uma_zdestroy(nfsmount_zone);
	return (0);
}

void 
nfs_dircookie_lock(struct nfsnode *np)
{
	mtx_lock(&np->n_mtx);
	while (np->n_flag & NDIRCOOKIELK)
		(void) msleep(&np->n_flag, &np->n_mtx, PZERO, "nfsdirlk", 0);
	np->n_flag |= NDIRCOOKIELK;
	mtx_unlock(&np->n_mtx);
}

void 
nfs_dircookie_unlock(struct nfsnode *np)
{
	mtx_lock(&np->n_mtx);
	np->n_flag &= ~NDIRCOOKIELK;
	wakeup(&np->n_flag);
	mtx_unlock(&np->n_mtx);
}

int
nfs_upgrade_vnlock(struct vnode *vp)
{
	int old_lock;

	ASSERT_VOP_LOCKED(vp, "nfs_upgrade_vnlock");
	old_lock = VOP_ISLOCKED(vp);
	if (old_lock != LK_EXCLUSIVE) {
		KASSERT(old_lock == LK_SHARED,
		    ("nfs_upgrade_vnlock: wrong old_lock %d", old_lock));
		/* Upgrade to exclusive lock, this might block */
		vn_lock(vp, LK_UPGRADE | LK_RETRY);
  	}
	return (old_lock);
}

void
nfs_downgrade_vnlock(struct vnode *vp, int old_lock)
{
	if (old_lock != LK_EXCLUSIVE) {
		KASSERT(old_lock == LK_SHARED, ("wrong old_lock %d", old_lock));
		/* Downgrade from exclusive lock. */
		vn_lock(vp, LK_DOWNGRADE | LK_RETRY);
  	}
}

void
nfs_printf(const char *fmt, ...)
{
	va_list ap;

	mtx_lock(&Giant);
	va_start(ap, fmt);
	printf(fmt, ap);
	va_end(ap);
	mtx_unlock(&Giant);
}

/*
 * Attribute cache routines.
 * nfs_loadattrcache() - loads or updates the cache contents from attributes
 *	that are on the mbuf list
 * nfs_getattrcache() - returns valid attributes if found in cache, returns
 *	error otherwise
 */

/*
 * Load the attribute cache (that lives in the nfsnode entry) with
 * the values on the mbuf list and
 * Iff vap not NULL
 *    copy the attributes to *vaper
 */
int
nfs_loadattrcache(struct vnode **vpp, struct mbuf **mdp, caddr_t *dposp,
		  struct vattr *vaper, int dontshrink)
{
	struct vnode *vp = *vpp;
	struct vattr *vap;
	struct nfs_fattr *fp;
	struct nfsnode *np = NULL;
	int32_t t1;
	caddr_t cp2;
	int rdev;
	struct mbuf *md;
	enum vtype vtyp;
	u_short vmode;
	struct timespec mtime, mtime_save;
	int v3 = NFS_ISV3(vp);
	struct thread *td = curthread;
	int error = 0;

	md = *mdp;
	t1 = (mtod(md, caddr_t) + md->m_len) - *dposp;
	cp2 = nfsm_disct(mdp, dposp, NFSX_FATTR(v3), t1, M_WAIT);
	if (cp2 == NULL) {
		error = EBADRPC;
		goto out;
	}
	fp = (struct nfs_fattr *)cp2;
	if (v3) {
		vtyp = nfsv3tov_type(fp->fa_type);
		vmode = fxdr_unsigned(u_short, fp->fa_mode);
		rdev = makedev(fxdr_unsigned(int, fp->fa3_rdev.specdata1),
			fxdr_unsigned(int, fp->fa3_rdev.specdata2));
		fxdr_nfsv3time(&fp->fa3_mtime, &mtime);
	} else {
		vtyp = nfsv2tov_type(fp->fa_type);
		vmode = fxdr_unsigned(u_short, fp->fa_mode);
		/*
		 * XXX
		 *
		 * The duplicate information returned in fa_type and fa_mode
		 * is an ambiguity in the NFS version 2 protocol.
		 *
		 * VREG should be taken literally as a regular file.  If a
		 * server intents to return some type information differently
		 * in the upper bits of the mode field (e.g. for sockets, or
		 * FIFOs), NFSv2 mandates fa_type to be VNON.  Anyway, we
		 * leave the examination of the mode bits even in the VREG
		 * case to avoid breakage for bogus servers, but we make sure
		 * that there are actually type bits set in the upper part of
		 * fa_mode (and failing that, trust the va_type field).
		 *
		 * NFSv3 cleared the issue, and requires fa_mode to not
		 * contain any type information (while also introduing sockets
		 * and FIFOs for fa_type).
		 */
		if (vtyp == VNON || (vtyp == VREG && (vmode & S_IFMT) != 0))
			vtyp = IFTOVT(vmode);
		rdev = fxdr_unsigned(int32_t, fp->fa2_rdev);
		fxdr_nfsv2time(&fp->fa2_mtime, &mtime);

		/*
		 * Really ugly NFSv2 kludge.
		 */
		if (vtyp == VCHR && rdev == 0xffffffff)
			vtyp = VFIFO;
	}

	/*
	 * If v_type == VNON it is a new node, so fill in the v_type,
	 * n_mtime fields. Check to see if it represents a special
	 * device, and if so, check for a possible alias. Once the
	 * correct vnode has been obtained, fill in the rest of the
	 * information.
	 */
	np = VTONFS(vp);
	mtx_lock(&np->n_mtx);
	if (vp->v_type != vtyp) {
		vp->v_type = vtyp;
		if (vp->v_type == VFIFO)
			vp->v_op = &nfs_fifoops;
		np->n_mtime = mtime;
	}
	vap = &np->n_vattr;
	vap->va_type = vtyp;
	vap->va_mode = (vmode & 07777);
	vap->va_rdev = rdev;
	mtime_save = vap->va_mtime;
	vap->va_mtime = mtime;
	vap->va_fsid = vp->v_mount->mnt_stat.f_fsid.val[0];
	if (v3) {
		vap->va_nlink = fxdr_unsigned(u_short, fp->fa_nlink);
		vap->va_uid = fxdr_unsigned(uid_t, fp->fa_uid);
		vap->va_gid = fxdr_unsigned(gid_t, fp->fa_gid);
		vap->va_size = fxdr_hyper(&fp->fa3_size);
		vap->va_blocksize = NFS_FABLKSIZE;
		vap->va_bytes = fxdr_hyper(&fp->fa3_used);
		vap->va_fileid = fxdr_unsigned(int32_t,
		    fp->fa3_fileid.nfsuquad[1]);
		fxdr_nfsv3time(&fp->fa3_atime, &vap->va_atime);
		fxdr_nfsv3time(&fp->fa3_ctime, &vap->va_ctime);
		vap->va_flags = 0;
		vap->va_filerev = 0;
	} else {
		vap->va_nlink = fxdr_unsigned(u_short, fp->fa_nlink);
		vap->va_uid = fxdr_unsigned(uid_t, fp->fa_uid);
		vap->va_gid = fxdr_unsigned(gid_t, fp->fa_gid);
		vap->va_size = fxdr_unsigned(u_int32_t, fp->fa2_size);
		vap->va_blocksize = fxdr_unsigned(int32_t, fp->fa2_blocksize);
		vap->va_bytes = (u_quad_t)fxdr_unsigned(int32_t, fp->fa2_blocks)
		    * NFS_FABLKSIZE;
		vap->va_fileid = fxdr_unsigned(int32_t, fp->fa2_fileid);
		fxdr_nfsv2time(&fp->fa2_atime, &vap->va_atime);
		vap->va_flags = 0;
		vap->va_ctime.tv_sec = fxdr_unsigned(u_int32_t,
		    fp->fa2_ctime.nfsv2_sec);
		vap->va_ctime.tv_nsec = 0;
		vap->va_gen = fxdr_unsigned(u_int32_t, fp->fa2_ctime.nfsv2_usec);
		vap->va_filerev = 0;
	}
	np->n_attrstamp = time_second;
	/* Timestamp the NFS otw getattr fetch */
	if (td->td_proc) {
		np->n_ac_ts_tid = td->td_tid;
		np->n_ac_ts_pid = td->td_proc->p_pid;
		np->n_ac_ts_syscalls = td->td_syscalls;
	} else
		bzero(&np->n_ac_ts, sizeof(struct nfs_attrcache_timestamp));
	
	if (vap->va_size != np->n_size) {
		if (vap->va_type == VREG) {
			if (dontshrink && vap->va_size < np->n_size) {
				/*
				 * We've been told not to shrink the file;
				 * zero np->n_attrstamp to indicate that
				 * the attributes are stale.
				 */
				vap->va_size = np->n_size;
				np->n_attrstamp = 0;
				KDTRACE_NFS_ATTRCACHE_FLUSH_DONE(vp);
			} else if (np->n_flag & NMODIFIED) {
				/*
				 * We've modified the file: Use the larger
				 * of our size, and the server's size.
				 */
				if (vap->va_size < np->n_size) {
					vap->va_size = np->n_size;
				} else {
					np->n_size = vap->va_size;
					np->n_flag |= NSIZECHANGED;
				}
			} else {
				np->n_size = vap->va_size;
				np->n_flag |= NSIZECHANGED;
			}
			vnode_pager_setsize(vp, np->n_size);
		} else {
			np->n_size = vap->va_size;
		}
	}
	/*
	 * The following checks are added to prevent a race between (say)
	 * a READDIR+ and a WRITE. 
	 * READDIR+, WRITE requests sent out.
	 * READDIR+ resp, WRITE resp received on client.
	 * However, the WRITE resp was handled before the READDIR+ resp
	 * causing the post op attrs from the write to be loaded first
	 * and the attrs from the READDIR+ to be loaded later. If this 
	 * happens, we have stale attrs loaded into the attrcache.
	 * We detect this by for the mtime moving back. We invalidate the 
	 * attrcache when this happens.
	 */
	if (timespeccmp(&mtime_save, &vap->va_mtime, >)) {
		/* Size changed or mtime went backwards */
		np->n_attrstamp = 0;
		KDTRACE_NFS_ATTRCACHE_FLUSH_DONE(vp);
	}
	if (vaper != NULL) {
		bcopy((caddr_t)vap, (caddr_t)vaper, sizeof(*vap));
		if (np->n_flag & NCHG) {
			if (np->n_flag & NACC)
				vaper->va_atime = np->n_atim;
			if (np->n_flag & NUPD)
				vaper->va_mtime = np->n_mtim;
		}
	}

#ifdef KDTRACE_HOOKS
	if (np->n_attrstamp != 0)
		KDTRACE_NFS_ATTRCACHE_LOAD_DONE(vp, &np->n_vattr, 0);
#endif
	mtx_unlock(&np->n_mtx);
out:
#ifdef KDTRACE_HOOKS
	if (error)
		KDTRACE_NFS_ATTRCACHE_LOAD_DONE(vp, NULL, error);
#endif
	return (error);
}

#ifdef NFS_ACDEBUG
#include <sys/sysctl.h>
SYSCTL_DECL(_vfs_nfs);
static int nfs_acdebug;
SYSCTL_INT(_vfs_nfs, OID_AUTO, acdebug, CTLFLAG_RW, &nfs_acdebug, 0,
    "Toggle acdebug (access cache debug) flag");
#endif

/*
 * Check the time stamp
 * If the cache is valid, copy contents to *vap and return 0
 * otherwise return an error
 */
int
nfs_getattrcache(struct vnode *vp, struct vattr *vaper)
{
	struct nfsnode *np;
	struct vattr *vap;
	struct nfsmount *nmp;
	int timeo;
	
	np = VTONFS(vp);
	vap = &np->n_vattr;
	nmp = VFSTONFS(vp->v_mount);
#ifdef NFS_ACDEBUG
	mtx_lock(&Giant);	/* nfs_printf() */
#endif
	mtx_lock(&np->n_mtx);
	/* XXX n_mtime doesn't seem to be updated on a miss-and-reload */
	timeo = (time_second - np->n_mtime.tv_sec) / 10;

#ifdef NFS_ACDEBUG
	if (nfs_acdebug>1)
		nfs_printf("nfs_getattrcache: initial timeo = %d\n", timeo);
#endif

	if (vap->va_type == VDIR) {
		if ((np->n_flag & NMODIFIED) || timeo < nmp->nm_acdirmin)
			timeo = nmp->nm_acdirmin;
		else if (timeo > nmp->nm_acdirmax)
			timeo = nmp->nm_acdirmax;
	} else {
		if ((np->n_flag & NMODIFIED) || timeo < nmp->nm_acregmin)
			timeo = nmp->nm_acregmin;
		else if (timeo > nmp->nm_acregmax)
			timeo = nmp->nm_acregmax;
	}

#ifdef NFS_ACDEBUG
	if (nfs_acdebug > 2)
		nfs_printf("acregmin %d; acregmax %d; acdirmin %d; acdirmax %d\n",
			   nmp->nm_acregmin, nmp->nm_acregmax,
			   nmp->nm_acdirmin, nmp->nm_acdirmax);

	if (nfs_acdebug)
		nfs_printf("nfs_getattrcache: age = %d; final timeo = %d\n",
			   (time_second - np->n_attrstamp), timeo);
#endif

	if ((time_second - np->n_attrstamp) >= timeo) {
		nfsstats.attrcache_misses++;
		mtx_unlock(&np->n_mtx);
		KDTRACE_NFS_ATTRCACHE_GET_MISS(vp);
		return (ENOENT);
	}
	nfsstats.attrcache_hits++;
	if (vap->va_size != np->n_size) {
		if (vap->va_type == VREG) {
			if (np->n_flag & NMODIFIED) {
				if (vap->va_size < np->n_size)
					vap->va_size = np->n_size;
				else
					np->n_size = vap->va_size;
			} else {
				np->n_size = vap->va_size;
			}
			vnode_pager_setsize(vp, np->n_size);
		} else {
			np->n_size = vap->va_size;
		}
	}
	bcopy((caddr_t)vap, (caddr_t)vaper, sizeof(struct vattr));
	if (np->n_flag & NCHG) {
		if (np->n_flag & NACC)
			vaper->va_atime = np->n_atim;
		if (np->n_flag & NUPD)
			vaper->va_mtime = np->n_mtim;
	}
	mtx_unlock(&np->n_mtx);
#ifdef NFS_ACDEBUG
	mtx_unlock(&Giant);	/* nfs_printf() */
#endif
	KDTRACE_NFS_ATTRCACHE_GET_HIT(vp, vap);
	return (0);
}

/*
 * Purge all cached information about an NFS vnode including name
 * cache entries, the attribute cache, and the access cache.  This is
 * called when an NFS request for a node fails with a stale
 * filehandle.
 */
void
nfs_purgecache(struct vnode *vp)
{
	struct nfsnode *np;
	int i;

	np = VTONFS(vp);
	cache_purge(vp);
	mtx_lock(&np->n_mtx);
	np->n_attrstamp = 0;
	KDTRACE_NFS_ATTRCACHE_FLUSH_DONE(vp);
	for (i = 0; i < NFS_ACCESSCACHESIZE; i++)
		np->n_accesscache[i].stamp = 0;
	KDTRACE_NFS_ACCESSCACHE_FLUSH_DONE(vp);
	mtx_unlock(&np->n_mtx);
}

static nfsuint64 nfs_nullcookie = { { 0, 0 } };
/*
 * This function finds the directory cookie that corresponds to the
 * logical byte offset given.
 */
nfsuint64 *
nfs_getcookie(struct nfsnode *np, off_t off, int add)
{
	struct nfsdmap *dp, *dp2;
	int pos;
	nfsuint64 *retval = NULL;
	
	pos = (uoff_t)off / NFS_DIRBLKSIZ;
	if (pos == 0 || off < 0) {
		KASSERT(!add, ("nfs getcookie add at <= 0"));
		return (&nfs_nullcookie);
	}
	pos--;
	dp = LIST_FIRST(&np->n_cookies);
	if (!dp) {
		if (add) {
			dp = malloc(sizeof (struct nfsdmap),
				M_NFSDIROFF, M_WAITOK);
			dp->ndm_eocookie = 0;
			LIST_INSERT_HEAD(&np->n_cookies, dp, ndm_list);
		} else
			goto out;
	}
	while (pos >= NFSNUMCOOKIES) {
		pos -= NFSNUMCOOKIES;
		if (LIST_NEXT(dp, ndm_list)) {
			if (!add && dp->ndm_eocookie < NFSNUMCOOKIES &&
			    pos >= dp->ndm_eocookie)
				goto out;
			dp = LIST_NEXT(dp, ndm_list);
		} else if (add) {
			dp2 = malloc(sizeof (struct nfsdmap),
				M_NFSDIROFF, M_WAITOK);
			dp2->ndm_eocookie = 0;
			LIST_INSERT_AFTER(dp, dp2, ndm_list);
			dp = dp2;
		} else
			goto out;
	}
	if (pos >= dp->ndm_eocookie) {
		if (add)
			dp->ndm_eocookie = pos + 1;
		else
			goto out;
	}
	retval = &dp->ndm_cookies[pos];
out:
	return (retval);
}

/*
 * Invalidate cached directory information, except for the actual directory
 * blocks (which are invalidated separately).
 * Done mainly to avoid the use of stale offset cookies.
 */
void
nfs_invaldir(struct vnode *vp)
{
	struct nfsnode *np = VTONFS(vp);

	KASSERT(vp->v_type == VDIR, ("nfs: invaldir not dir"));
	nfs_dircookie_lock(np);
	np->n_direofoffset = 0;
	np->n_cookieverf.nfsuquad[0] = 0;
	np->n_cookieverf.nfsuquad[1] = 0;
	if (LIST_FIRST(&np->n_cookies))
		LIST_FIRST(&np->n_cookies)->ndm_eocookie = 0;
	nfs_dircookie_unlock(np);
}

/*
 * The write verifier has changed (probably due to a server reboot), so all
 * B_NEEDCOMMIT blocks will have to be written again. Since they are on the
 * dirty block list as B_DELWRI, all this takes is clearing the B_NEEDCOMMIT
 * and B_CLUSTEROK flags.  Once done the new write verifier can be set for the
 * mount point.
 *
 * B_CLUSTEROK must be cleared along with B_NEEDCOMMIT because stage 1 data
 * writes are not clusterable.
 */
void
nfs_clearcommit(struct mount *mp)
{
	struct vnode *vp, *nvp;
	struct buf *bp, *nbp;
	struct bufobj *bo;

	MNT_ILOCK(mp);
	MNT_VNODE_FOREACH(vp, mp, nvp) {
		bo = &vp->v_bufobj;
		VI_LOCK(vp);
		if (vp->v_iflag & VI_DOOMED) {
			VI_UNLOCK(vp);
			continue;
		}
		vholdl(vp);
		VI_UNLOCK(vp);
		MNT_IUNLOCK(mp);
		BO_LOCK(bo);
		TAILQ_FOREACH_SAFE(bp, &bo->bo_dirty.bv_hd, b_bobufs, nbp) {
			if (!BUF_ISLOCKED(bp) &&
			    (bp->b_flags & (B_DELWRI | B_NEEDCOMMIT))
				== (B_DELWRI | B_NEEDCOMMIT))
				bp->b_flags &= ~(B_NEEDCOMMIT | B_CLUSTEROK);
		}
		BO_UNLOCK(bo);
		vdrop(vp);
		MNT_ILOCK(mp);
	}
	MNT_IUNLOCK(mp);
}

/*
 * Helper functions for former macros.  Some of these should be
 * moved to their callers.
 */

int
nfsm_mtofh_xx(struct vnode *d, struct vnode **v, int v3, int *f,
    struct mbuf **md, caddr_t *dpos)
{
	struct nfsnode *ttnp;
	struct vnode *ttvp;
	nfsfh_t *ttfhp;
	u_int32_t *tl;
	int ttfhsize;
	int t1;

	if (v3) {
		tl = nfsm_dissect_xx(NFSX_UNSIGNED, md, dpos);
		if (tl == NULL)
			return EBADRPC;
		*f = fxdr_unsigned(int, *tl);
	} else
		*f = 1;
	if (*f) {
		t1 = nfsm_getfh_xx(&ttfhp, &ttfhsize, (v3), md, dpos);
		if (t1 != 0)
			return t1;
		t1 = nfs_nget(d->v_mount, ttfhp, ttfhsize, &ttnp, LK_EXCLUSIVE);
		if (t1 != 0)
			return t1;
		*v = NFSTOV(ttnp);
	}
	if (v3) {
		tl = nfsm_dissect_xx(NFSX_UNSIGNED, md, dpos);
		if (tl == NULL)
			return EBADRPC;
		if (*f)
			*f = fxdr_unsigned(int, *tl);
		else if (fxdr_unsigned(int, *tl))
			nfsm_adv_xx(NFSX_V3FATTR, md, dpos);
	}
	if (*f) {
		ttvp = *v;
		t1 = nfs_loadattrcache(&ttvp, md, dpos, NULL, 0);
		if (t1)
			return t1;
		*v = ttvp;
	}
	return 0;
}

int
nfsm_getfh_xx(nfsfh_t **f, int *s, int v3, struct mbuf **md, caddr_t *dpos)
{
	u_int32_t *tl;

	if (v3) {
		tl = nfsm_dissect_xx(NFSX_UNSIGNED, md, dpos);
		if (tl == NULL)
			return EBADRPC;
		*s = fxdr_unsigned(int, *tl);
		if (*s <= 0 || *s > NFSX_V3FHMAX)
			return EBADRPC;
	} else
		*s = NFSX_V2FH;
	*f = nfsm_dissect_xx(nfsm_rndup(*s), md, dpos);
	if (*f == NULL)
		return EBADRPC;
	else
		return 0;
}


int
nfsm_loadattr_xx(struct vnode **v, struct vattr *va, struct mbuf **md,
		 caddr_t *dpos)
{
	int t1;

	struct vnode *ttvp = *v;
	t1 = nfs_loadattrcache(&ttvp, md, dpos, va, 0);
	if (t1 != 0)
		return t1;
	*v = ttvp;
	return 0;
}

int
nfsm_postop_attr_xx(struct vnode **v, int *f, struct mbuf **md,
		    caddr_t *dpos)
{
	u_int32_t *tl;
	int t1;

	struct vnode *ttvp = *v;
	tl = nfsm_dissect_xx(NFSX_UNSIGNED, md, dpos);
	if (tl == NULL)
		return EBADRPC;
	*f = fxdr_unsigned(int, *tl);
	if (*f != 0) {
		t1 = nfs_loadattrcache(&ttvp, md, dpos, NULL, 1);
		if (t1 != 0) {
			*f = 0;
			return t1;
		}
		*v = ttvp;
	}
	return 0;
}

int
nfsm_wcc_data_xx(struct vnode **v, int *f, struct mbuf **md, caddr_t *dpos)
{
	u_int32_t *tl;
	int ttattrf, ttretf = 0;
	int t1;

	tl = nfsm_dissect_xx(NFSX_UNSIGNED, md, dpos);
	if (tl == NULL)
		return EBADRPC;
	if (*tl == nfs_true) {
		tl = nfsm_dissect_xx(6 * NFSX_UNSIGNED, md, dpos);
		if (tl == NULL)
			return EBADRPC;
		mtx_lock(&(VTONFS(*v))->n_mtx);
		if (*f)
 			ttretf = (VTONFS(*v)->n_mtime.tv_sec == fxdr_unsigned(u_int32_t, *(tl + 2)) && 
				  VTONFS(*v)->n_mtime.tv_nsec == fxdr_unsigned(u_int32_t, *(tl + 3))); 
		mtx_unlock(&(VTONFS(*v))->n_mtx);
	}
	t1 = nfsm_postop_attr_xx(v, &ttattrf, md, dpos);
	if (t1)
		return t1;
	if (*f)
		*f = ttretf;
	else
		*f = ttattrf;
	return 0;
}

int
nfsm_strtom_xx(const char *a, int s, int m, struct mbuf **mb, caddr_t *bpos)
{
	u_int32_t *tl;
	int t1;

	if (s > m)
		return ENAMETOOLONG;
	t1 = nfsm_rndup(s) + NFSX_UNSIGNED;
	if (t1 <= M_TRAILINGSPACE(*mb)) {
		tl = nfsm_build_xx(t1, mb, bpos);
		*tl++ = txdr_unsigned(s);
		*(tl + ((t1 >> 2) - 2)) = 0;
		bcopy(a, tl, s);
	} else {
		t1 = nfsm_strtmbuf(mb, bpos, a, s);
		if (t1 != 0)
			return t1;
	}
	return 0;
}

int
nfsm_fhtom_xx(struct vnode *v, int v3, struct mbuf **mb, caddr_t *bpos)
{
	u_int32_t *tl;
	int t1;
	caddr_t cp;

	if (v3) {
		t1 = nfsm_rndup(VTONFS(v)->n_fhsize) + NFSX_UNSIGNED;
		if (t1 < M_TRAILINGSPACE(*mb)) {
			tl = nfsm_build_xx(t1, mb, bpos);
			*tl++ = txdr_unsigned(VTONFS(v)->n_fhsize);
			*(tl + ((t1 >> 2) - 2)) = 0;
			bcopy(VTONFS(v)->n_fhp, tl, VTONFS(v)->n_fhsize);
		} else {
			t1 = nfsm_strtmbuf(mb, bpos,
			    (const char *)VTONFS(v)->n_fhp,
			    VTONFS(v)->n_fhsize);
			if (t1 != 0)
				return t1;
		}
	} else {
		cp = nfsm_build_xx(NFSX_V2FH, mb, bpos);
		bcopy(VTONFS(v)->n_fhp, cp, NFSX_V2FH);
	}
	return 0;
}

void
nfsm_v3attrbuild_xx(struct vattr *va, int full, struct mbuf **mb,
    caddr_t *bpos)
{
	u_int32_t *tl;

	if (va->va_mode != (mode_t)VNOVAL) {
		tl = nfsm_build_xx(2 * NFSX_UNSIGNED, mb, bpos);
		*tl++ = nfs_true;
		*tl = txdr_unsigned(va->va_mode);
	} else {
		tl = nfsm_build_xx(NFSX_UNSIGNED, mb, bpos);
		*tl = nfs_false;
	}
	if (full && va->va_uid != (uid_t)VNOVAL) {
		tl = nfsm_build_xx(2 * NFSX_UNSIGNED, mb, bpos);
		*tl++ = nfs_true;
		*tl = txdr_unsigned(va->va_uid);
	} else {
		tl = nfsm_build_xx(NFSX_UNSIGNED, mb, bpos);
		*tl = nfs_false;
	}
	if (full && va->va_gid != (gid_t)VNOVAL) {
		tl = nfsm_build_xx(2 * NFSX_UNSIGNED, mb, bpos);
		*tl++ = nfs_true;
		*tl = txdr_unsigned(va->va_gid);
	} else {
		tl = nfsm_build_xx(NFSX_UNSIGNED, mb, bpos);
		*tl = nfs_false;
	}
	if (full && va->va_size != VNOVAL) {
		tl = nfsm_build_xx(3 * NFSX_UNSIGNED, mb, bpos);
		*tl++ = nfs_true;
		txdr_hyper(va->va_size, tl);
	} else {
		tl = nfsm_build_xx(NFSX_UNSIGNED, mb, bpos);
		*tl = nfs_false;
	}
	if (va->va_atime.tv_sec != VNOVAL) {
		if (va->va_atime.tv_sec != time_second) {
			tl = nfsm_build_xx(3 * NFSX_UNSIGNED, mb, bpos);
			*tl++ = txdr_unsigned(NFSV3SATTRTIME_TOCLIENT);
			txdr_nfsv3time(&va->va_atime, tl);
		} else {
			tl = nfsm_build_xx(NFSX_UNSIGNED, mb, bpos);
			*tl = txdr_unsigned(NFSV3SATTRTIME_TOSERVER);
		}
	} else {
		tl = nfsm_build_xx(NFSX_UNSIGNED, mb, bpos);
		*tl = txdr_unsigned(NFSV3SATTRTIME_DONTCHANGE);
	}
	if (va->va_mtime.tv_sec != VNOVAL) {
		if (va->va_mtime.tv_sec != time_second) {
			tl = nfsm_build_xx(3 * NFSX_UNSIGNED, mb, bpos);
			*tl++ = txdr_unsigned(NFSV3SATTRTIME_TOCLIENT);
			txdr_nfsv3time(&va->va_mtime, tl);
		} else {
			tl = nfsm_build_xx(NFSX_UNSIGNED, mb, bpos);
			*tl = txdr_unsigned(NFSV3SATTRTIME_TOSERVER);
		}
	} else {
		tl = nfsm_build_xx(NFSX_UNSIGNED, mb, bpos);
		*tl = txdr_unsigned(NFSV3SATTRTIME_DONTCHANGE);
	}
}
OpenPOWER on IntegriCloud