summaryrefslogtreecommitdiffstats
path: root/sys/kern/vfs_vnops.c
blob: 7891dc6b486a3277ea3fc42c4702b5f803d03150 (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
1147
1148
1149
1150
1151
/*
 * Copyright (c) 1982, 1986, 1989, 1993
 *	The Regents of the University of California.  All rights reserved.
 * (c) UNIX System Laboratories, Inc.
 * All or some portions of this file are derived from material licensed
 * to the University of California by American Telephone and Telegraph
 * Co. or Unix System Laboratories, Inc. and are reproduced herein with
 * the permission of UNIX System Laboratories, Inc.
 *
 * 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.
 * 3. All advertising materials mentioning features or use of this software
 *    must display the following acknowledgement:
 *	This product includes software developed by the University of
 *	California, Berkeley and its contributors.
 * 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.
 *
 *	@(#)vfs_vnops.c	8.2 (Berkeley) 1/21/94
 * $FreeBSD$
 */

#include "opt_mac.h"

#include <sys/param.h>
#include <sys/systm.h>
#include <sys/fcntl.h>
#include <sys/file.h>
#include <sys/stat.h>
#include <sys/proc.h>
#include <sys/lock.h>
#include <sys/mac.h>
#include <sys/mount.h>
#include <sys/mutex.h>
#include <sys/namei.h>
#include <sys/vnode.h>
#include <sys/bio.h>
#include <sys/buf.h>
#include <sys/filio.h>
#include <sys/sx.h>
#include <sys/ttycom.h>
#include <sys/conf.h>
#include <sys/syslog.h>

#include <machine/limits.h>

static fo_rdwr_t	vn_read;
static fo_rdwr_t	vn_write;
static fo_ioctl_t	vn_ioctl;
static fo_poll_t	vn_poll;
static fo_kqfilter_t	vn_kqfilter;
static fo_stat_t	vn_statfile;
static fo_close_t	vn_closefile;

struct 	fileops vnops = {
	vn_read, vn_write, vn_ioctl, vn_poll, vn_kqfilter,
	vn_statfile, vn_closefile
};

int
vn_open(ndp, flagp, cmode)
	register struct nameidata *ndp;
	int *flagp, cmode;
{
	struct thread *td = ndp->ni_cnd.cn_thread;

	return (vn_open_cred(ndp, flagp, cmode, td->td_ucred));
}

/*
 * Common code for vnode open operations.
 * Check permissions, and call the VOP_OPEN or VOP_CREATE routine.
 * 
 * Note that this does NOT free nameidata for the successful case,
 * due to the NDINIT being done elsewhere.
 */
int
vn_open_cred(ndp, flagp, cmode, cred)
	register struct nameidata *ndp;
	int *flagp, cmode;
	struct ucred *cred;
{
	struct vnode *vp;
	struct mount *mp;
	struct thread *td = ndp->ni_cnd.cn_thread;
	struct vattr vat;
	struct vattr *vap = &vat;
	int mode, fmode, error;
#ifdef LOOKUP_SHARED
	int exclusive;	/* The current intended lock state */

	exclusive = 0;
#endif

restart:
	fmode = *flagp;
	if (fmode & O_CREAT) {
		ndp->ni_cnd.cn_nameiop = CREATE;
		ndp->ni_cnd.cn_flags = LOCKPARENT | LOCKLEAF;
		if ((fmode & O_EXCL) == 0 && (fmode & O_NOFOLLOW) == 0)
			ndp->ni_cnd.cn_flags |= FOLLOW;
		bwillwrite();
		if ((error = namei(ndp)) != 0)
			return (error);
		if (ndp->ni_vp == NULL) {
			VATTR_NULL(vap);
			vap->va_type = VREG;
			vap->va_mode = cmode;
			if (fmode & O_EXCL)
				vap->va_vaflags |= VA_EXCLUSIVE;
			if (vn_start_write(ndp->ni_dvp, &mp, V_NOWAIT) != 0) {
				NDFREE(ndp, NDF_ONLY_PNBUF);
				vput(ndp->ni_dvp);
				if ((error = vn_start_write(NULL, &mp,
				    V_XSLEEP | PCATCH)) != 0)
					return (error);
				goto restart;
			}
#ifdef MAC
			error = mac_check_vnode_create(cred, ndp->ni_dvp,
			    &ndp->ni_cnd, vap);
			if (error == 0) {
#endif
				VOP_LEASE(ndp->ni_dvp, td, cred, LEASE_WRITE);
				error = VOP_CREATE(ndp->ni_dvp, &ndp->ni_vp,
						   &ndp->ni_cnd, vap);
#ifdef MAC
			}
#endif
			vput(ndp->ni_dvp);
			vn_finished_write(mp);
			if (error) {
				NDFREE(ndp, NDF_ONLY_PNBUF);
				return (error);
			}
			ASSERT_VOP_UNLOCKED(ndp->ni_dvp, "create");
			ASSERT_VOP_LOCKED(ndp->ni_vp, "create");
			fmode &= ~O_TRUNC;
			vp = ndp->ni_vp;
#ifdef LOOKUP_SHARED
			exclusive = 1;
#endif
		} else {
			if (ndp->ni_dvp == ndp->ni_vp)
				vrele(ndp->ni_dvp);
			else
				vput(ndp->ni_dvp);
			ndp->ni_dvp = NULL;
			vp = ndp->ni_vp;
			if (fmode & O_EXCL) {
				error = EEXIST;
				goto bad;
			}
			fmode &= ~O_CREAT;
		}
	} else {
		ndp->ni_cnd.cn_nameiop = LOOKUP;
#ifdef LOOKUP_SHARED
		ndp->ni_cnd.cn_flags =
		    ((fmode & O_NOFOLLOW) ? NOFOLLOW : FOLLOW) |
		    LOCKSHARED | LOCKLEAF;
#else
		ndp->ni_cnd.cn_flags =
		    ((fmode & O_NOFOLLOW) ? NOFOLLOW : FOLLOW) | LOCKLEAF;
#endif
		if ((error = namei(ndp)) != 0)
			return (error);
		vp = ndp->ni_vp;
	}
	if (vp->v_type == VLNK) {
		error = EMLINK;
		goto bad;
	}
	if (vp->v_type == VSOCK) {
		error = EOPNOTSUPP;
		goto bad;
	}
	mode = 0;
	if (fmode & (FWRITE | O_TRUNC)) {
		if (vp->v_type == VDIR) {
			error = EISDIR;
			goto bad;
		}
		mode |= VWRITE;
	}
	if (fmode & FREAD)
		mode |= VREAD;
	if (fmode & O_APPEND)
		mode |= VAPPEND;
#ifdef MAC
	error = mac_check_vnode_open(cred, vp, mode);
	if (error)
		goto bad;
#endif
	if ((fmode & O_CREAT) == 0) {
		if (mode & VWRITE) {
			error = vn_writechk(vp);
			if (error)
				goto bad;
		}
		if (mode) {
		        error = VOP_ACCESS(vp, mode, cred, td);
			if (error)
				goto bad;
		}
	}
	if ((error = VOP_GETATTR(vp, vap, cred, td)) == 0) {
		vp->v_cachedfs = vap->va_fsid;
		vp->v_cachedid = vap->va_fileid;
	}
	if ((error = VOP_OPEN(vp, fmode, cred, td)) != 0)
		goto bad;
	/*
	 * Make sure that a VM object is created for VMIO support.
	 */
	if (vn_canvmio(vp) == TRUE) {
#ifdef LOOKUP_SHARED
		int flock;

		if (!exclusive && VOP_GETVOBJECT(vp, NULL) != 0)
			VOP_LOCK(vp, LK_UPGRADE, td);
		/*
		 * In cases where the object is marked as dead object_create
		 * will unlock and relock exclusive.  It is safe to call in
		 * here with a shared lock because we only examine fields that
		 * the shared lock guarantees will be stable.  In the UPGRADE
		 * case it is not likely that anyone has used this vnode yet
		 * so there will be no contention.  The logic after this call
		 * restores the requested locking state.
		 */
#endif
		if ((error = vfs_object_create(vp, td, cred)) != 0) {
			VOP_UNLOCK(vp, 0, td);
			VOP_CLOSE(vp, fmode, cred, td);
			NDFREE(ndp, NDF_ONLY_PNBUF);
			vrele(vp);
			*flagp = fmode;
			return (error);
		}
#ifdef LOOKUP_SHARED
		flock = VOP_ISLOCKED(vp, td);
		if (!exclusive && flock == LK_EXCLUSIVE)
			VOP_LOCK(vp, LK_DOWNGRADE, td);
#endif
	}

	if (fmode & FWRITE)
		vp->v_writecount++;
	*flagp = fmode;
	return (0);
bad:
	NDFREE(ndp, NDF_ONLY_PNBUF);
	vput(vp);
	*flagp = fmode;
	ndp->ni_vp = NULL;
	return (error);
}

/*
 * Check for write permissions on the specified vnode.
 * Prototype text segments cannot be written.
 */
int
vn_writechk(vp)
	register struct vnode *vp;
{

	ASSERT_VOP_LOCKED(vp, "vn_writechk");
	/*
	 * If there's shared text associated with
	 * the vnode, try to free it up once.  If
	 * we fail, we can't allow writing.
	 */
	if (vp->v_vflag & VV_TEXT)
		return (ETXTBSY);

	return (0);
}

/*
 * Vnode close call
 */
int
vn_close(vp, flags, file_cred, td)
	register struct vnode *vp;
	int flags;
	struct ucred *file_cred;
	struct thread *td;
{
	int error;

	if (flags & FWRITE)
		vp->v_writecount--;
	error = VOP_CLOSE(vp, flags, file_cred, td);
	/*
	 * XXX - In certain instances VOP_CLOSE has to do the vrele
	 * itself. If the vrele has been done, it will return EAGAIN
	 * to indicate that the vrele should not be done again. When
	 * this happens, we just return success. The correct thing to
	 * do would be to have all VOP_CLOSE instances do the vrele.
	 */
	if (error == EAGAIN)
		return (0);
	vrele(vp);
	return (error);
}

/*
 * Sequential heuristic - detect sequential operation
 */
static __inline
int
sequential_heuristic(struct uio *uio, struct file *fp)
{

	if ((uio->uio_offset == 0 && fp->f_seqcount > 0) ||
	    uio->uio_offset == fp->f_nextoff) {
		/*
		 * XXX we assume that the filesystem block size is
		 * the default.  Not true, but still gives us a pretty
		 * good indicator of how sequential the read operations
		 * are.
		 */
		fp->f_seqcount += (uio->uio_resid + BKVASIZE - 1) / BKVASIZE;
		if (fp->f_seqcount > IO_SEQMAX)
			fp->f_seqcount = IO_SEQMAX;
		return(fp->f_seqcount << IO_SEQSHIFT);
	}

	/*
	 * Not sequential, quick draw-down of seqcount
	 */
	if (fp->f_seqcount > 1)
		fp->f_seqcount = 1;
	else
		fp->f_seqcount = 0;
	return(0);
}

/*
 * Package up an I/O request on a vnode into a uio and do it.
 */
int
vn_rdwr(rw, vp, base, len, offset, segflg, ioflg, active_cred, file_cred,
    aresid, td)
	enum uio_rw rw;
	struct vnode *vp;
	caddr_t base;
	int len;
	off_t offset;
	enum uio_seg segflg;
	int ioflg;
	struct ucred *active_cred;
	struct ucred *file_cred;
	int *aresid;
	struct thread *td;
{
	struct uio auio;
	struct iovec aiov;
	struct mount *mp;
	struct ucred *cred;
	int error;

	if ((ioflg & IO_NODELOCKED) == 0) {
		mp = NULL;
		if (rw == UIO_WRITE) { 
			if (vp->v_type != VCHR &&
			    (error = vn_start_write(vp, &mp, V_WAIT | PCATCH))
			    != 0)
				return (error);
			vn_lock(vp, LK_EXCLUSIVE | LK_RETRY, td);
		} else {
			/*
			 * XXX This should be LK_SHARED but I don't trust VFS
			 * enough to leave it like that until it has been
			 * reviewed further.
			 */
			vn_lock(vp, LK_EXCLUSIVE | LK_RETRY, td);
		}

	}
	auio.uio_iov = &aiov;
	auio.uio_iovcnt = 1;
	aiov.iov_base = base;
	aiov.iov_len = len;
	auio.uio_resid = len;
	auio.uio_offset = offset;
	auio.uio_segflg = segflg;
	auio.uio_rw = rw;
	auio.uio_td = td;
	error = 0;
#ifdef MAC
	if ((ioflg & IO_NOMACCHECK) == 0) {
		if (rw == UIO_READ)
			error = mac_check_vnode_read(active_cred, file_cred,
			    vp);
		else
			error = mac_check_vnode_write(active_cred, file_cred,
			    vp);
	}
#endif
	if (error == 0) {
		if (file_cred)
			cred = file_cred;
		else
			cred = active_cred;
		if (rw == UIO_READ)
			error = VOP_READ(vp, &auio, ioflg, cred);
		else
			error = VOP_WRITE(vp, &auio, ioflg, cred);
	}
	if (aresid)
		*aresid = auio.uio_resid;
	else
		if (auio.uio_resid && error == 0)
			error = EIO;
	if ((ioflg & IO_NODELOCKED) == 0) {
		if (rw == UIO_WRITE)
			vn_finished_write(mp);
		VOP_UNLOCK(vp, 0, td);
	}
	return (error);
}

/*
 * Package up an I/O request on a vnode into a uio and do it.  The I/O
 * request is split up into smaller chunks and we try to avoid saturating
 * the buffer cache while potentially holding a vnode locked, so we 
 * check bwillwrite() before calling vn_rdwr().  We also call uio_yield()
 * to give other processes a chance to lock the vnode (either other processes
 * core'ing the same binary, or unrelated processes scanning the directory).
 */
int
vn_rdwr_inchunks(rw, vp, base, len, offset, segflg, ioflg, active_cred,
    file_cred, aresid, td)
	enum uio_rw rw;
	struct vnode *vp;
	caddr_t base;
	int len;
	off_t offset;
	enum uio_seg segflg;
	int ioflg;
	struct ucred *active_cred;
	struct ucred *file_cred;
	int *aresid;
	struct thread *td;
{
	int error = 0;

	do {
		int chunk = (len > MAXBSIZE) ? MAXBSIZE : len;

		if (rw != UIO_READ && vp->v_type == VREG)
			bwillwrite();
		error = vn_rdwr(rw, vp, base, chunk, offset, segflg,
		    ioflg, active_cred, file_cred, aresid, td);
		len -= chunk;	/* aresid calc already includes length */
		if (error)
			break;
		offset += chunk;
		base += chunk;
		uio_yield();
	} while (len);
	if (aresid)
		*aresid += len;
	return (error);
}

/*
 * File table vnode read routine.
 */
static int
vn_read(fp, uio, active_cred, flags, td)
	struct file *fp;
	struct uio *uio;
	struct ucred *active_cred;
	struct thread *td;
	int flags;
{
	struct vnode *vp;
	int error, ioflag;

	mtx_lock(&Giant);
	KASSERT(uio->uio_td == td, ("uio_td %p is not td %p",
	    uio->uio_td, td));
	vp = fp->f_data;
	ioflag = 0;
	if (fp->f_flag & FNONBLOCK)
		ioflag |= IO_NDELAY;
	if (fp->f_flag & O_DIRECT)
		ioflag |= IO_DIRECT;
	VOP_LEASE(vp, td, fp->f_cred, LEASE_READ);
	/*
	 * According to McKusick the vn lock is protecting f_offset here.
	 * Once this field has it's own lock we can acquire this shared.
	 */
	vn_lock(vp, LK_EXCLUSIVE | LK_NOPAUSE | LK_RETRY, td);
	if ((flags & FOF_OFFSET) == 0)
		uio->uio_offset = fp->f_offset;

	ioflag |= sequential_heuristic(uio, fp);

#ifdef MAC
	error = mac_check_vnode_read(active_cred, fp->f_cred, vp);
	if (error == 0)
#endif
		error = VOP_READ(vp, uio, ioflag, fp->f_cred);
	if ((flags & FOF_OFFSET) == 0)
		fp->f_offset = uio->uio_offset;
	fp->f_nextoff = uio->uio_offset;
	VOP_UNLOCK(vp, 0, td);
	mtx_unlock(&Giant);
	return (error);
}

/*
 * File table vnode write routine.
 */
static int
vn_write(fp, uio, active_cred, flags, td)
	struct file *fp;
	struct uio *uio;
	struct ucred *active_cred;
	struct thread *td;
	int flags;
{
	struct vnode *vp;
	struct mount *mp;
	int error, ioflag;

	mtx_lock(&Giant);
	KASSERT(uio->uio_td == td, ("uio_td %p is not td %p",
	    uio->uio_td, td));
	vp = fp->f_data;
	if (vp->v_type == VREG)
		bwillwrite();
	ioflag = IO_UNIT;
	if (vp->v_type == VREG && (fp->f_flag & O_APPEND))
		ioflag |= IO_APPEND;
	if (fp->f_flag & FNONBLOCK)
		ioflag |= IO_NDELAY;
	if (fp->f_flag & O_DIRECT)
		ioflag |= IO_DIRECT;
	if ((fp->f_flag & O_FSYNC) ||
	    (vp->v_mount && (vp->v_mount->mnt_flag & MNT_SYNCHRONOUS)))
		ioflag |= IO_SYNC;
	mp = NULL;
	if (vp->v_type != VCHR &&
	    (error = vn_start_write(vp, &mp, V_WAIT | PCATCH)) != 0) {
		mtx_unlock(&Giant);
		return (error);
	}
	VOP_LEASE(vp, td, fp->f_cred, LEASE_WRITE);
	vn_lock(vp, LK_EXCLUSIVE | LK_RETRY, td);
	if ((flags & FOF_OFFSET) == 0)
		uio->uio_offset = fp->f_offset;
	ioflag |= sequential_heuristic(uio, fp);
#ifdef MAC
	error = mac_check_vnode_write(active_cred, fp->f_cred, vp);
	if (error == 0)
#endif
		error = VOP_WRITE(vp, uio, ioflag, fp->f_cred);
	if ((flags & FOF_OFFSET) == 0)
		fp->f_offset = uio->uio_offset;
	fp->f_nextoff = uio->uio_offset;
	VOP_UNLOCK(vp, 0, td);
	vn_finished_write(mp);
	mtx_unlock(&Giant);
	return (error);
}

/*
 * File table vnode stat routine.
 */
static int
vn_statfile(fp, sb, active_cred, td)
	struct file *fp;
	struct stat *sb;
	struct ucred *active_cred;
	struct thread *td;
{
	struct vnode *vp = fp->f_data;
	int error;

	vn_lock(vp, LK_EXCLUSIVE | LK_RETRY, td);
	error = vn_stat(vp, sb, active_cred, fp->f_cred, td);
	VOP_UNLOCK(vp, 0, td);

	return (error);
}

/*
 * Stat a vnode; implementation for the stat syscall
 */
int
vn_stat(vp, sb, active_cred, file_cred, td)
	struct vnode *vp;
	register struct stat *sb;
	struct ucred *active_cred;
	struct ucred *file_cred;
	struct thread *td;
{
	struct vattr vattr;
	register struct vattr *vap;
	int error;
	u_short mode;

#ifdef MAC
	error = mac_check_vnode_stat(active_cred, file_cred, vp);
	if (error)
		return (error);
#endif

	vap = &vattr;
	error = VOP_GETATTR(vp, vap, active_cred, td);
	if (error)
		return (error);

	vp->v_cachedfs = vap->va_fsid;
	vp->v_cachedid = vap->va_fileid;

	/*
	 * Zero the spare stat fields
	 */
	bzero(sb, sizeof *sb);

	/*
	 * Copy from vattr table
	 */
	if (vap->va_fsid != VNOVAL)
		sb->st_dev = vap->va_fsid;
	else
		sb->st_dev = vp->v_mount->mnt_stat.f_fsid.val[0];
	sb->st_ino = vap->va_fileid;
	mode = vap->va_mode;
	switch (vap->va_type) {
	case VREG:
		mode |= S_IFREG;
		break;
	case VDIR:
		mode |= S_IFDIR;
		break;
	case VBLK:
		mode |= S_IFBLK;
		break;
	case VCHR:
		mode |= S_IFCHR;
		break;
	case VLNK:
		mode |= S_IFLNK;
		/* This is a cosmetic change, symlinks do not have a mode. */
		if (vp->v_mount->mnt_flag & MNT_NOSYMFOLLOW)
			sb->st_mode &= ~ACCESSPERMS;	/* 0000 */
		else
			sb->st_mode |= ACCESSPERMS;	/* 0777 */
		break;
	case VSOCK:
		mode |= S_IFSOCK;
		break;
	case VFIFO:
		mode |= S_IFIFO;
		break;
	default:
		return (EBADF);
	};
	sb->st_mode = mode;
	sb->st_nlink = vap->va_nlink;
	sb->st_uid = vap->va_uid;
	sb->st_gid = vap->va_gid;
	sb->st_rdev = vap->va_rdev;
	if (vap->va_size > OFF_MAX)
		return (EOVERFLOW);
	sb->st_size = vap->va_size;
	sb->st_atimespec = vap->va_atime;
	sb->st_mtimespec = vap->va_mtime;
	sb->st_ctimespec = vap->va_ctime;
	sb->st_birthtimespec = vap->va_birthtime;

        /*
	 * According to www.opengroup.org, the meaning of st_blksize is 
	 *   "a filesystem-specific preferred I/O block size for this 
	 *    object.  In some filesystem types, this may vary from file
	 *    to file"
	 * Default to PAGE_SIZE after much discussion.
	 */

	if (vap->va_type == VREG) {
		sb->st_blksize = vap->va_blocksize;
	} else if (vn_isdisk(vp, NULL)) {
		sb->st_blksize = vp->v_rdev->si_bsize_best;
		if (sb->st_blksize < vp->v_rdev->si_bsize_phys)
			sb->st_blksize = vp->v_rdev->si_bsize_phys;
		if (sb->st_blksize < BLKDEV_IOSIZE)
			sb->st_blksize = BLKDEV_IOSIZE;
	} else {
		sb->st_blksize = PAGE_SIZE;
	}
	
	sb->st_flags = vap->va_flags;
	if (suser(td))
		sb->st_gen = 0;
	else
		sb->st_gen = vap->va_gen;

#if (S_BLKSIZE == 512)
	/* Optimize this case */
	sb->st_blocks = vap->va_bytes >> 9;
#else
	sb->st_blocks = vap->va_bytes / S_BLKSIZE;
#endif
	return (0);
}

/*
 * File table vnode ioctl routine.
 */
static int
vn_ioctl(fp, com, data, active_cred, td)
	struct file *fp;
	u_long com;
	void *data;
	struct ucred *active_cred;
	struct thread *td;
{
	struct vnode *vp = fp->f_data;
	struct vnode *vpold;
	struct vattr vattr;
	int error;

	switch (vp->v_type) {

	case VREG:
	case VDIR:
		if (com == FIONREAD) {
			vn_lock(vp, LK_EXCLUSIVE | LK_RETRY, td);
			error = VOP_GETATTR(vp, &vattr, active_cred, td);
			VOP_UNLOCK(vp, 0, td);
			if (error)
				return (error);
			*(int *)data = vattr.va_size - fp->f_offset;
			return (0);
		}
		if (com == FIONBIO || com == FIOASYNC)	/* XXX */
			return (0);			/* XXX */
		/* FALLTHROUGH */

	default:
#if 0
		return (ENOTTY);
#endif
	case VFIFO:
	case VCHR:
	case VBLK:
		if (com == FIODTYPE) {
			if (vp->v_type != VCHR && vp->v_type != VBLK)
				return (ENOTTY);
			*(int *)data = devsw(vp->v_rdev)->d_flags & D_TYPEMASK;
			return (0);
		}
		error = VOP_IOCTL(vp, com, data, fp->f_flag, active_cred, td);
		if (error == ENOIOCTL) {
#ifdef DIAGNOSTIC
			Debugger("ENOIOCTL leaked through");
#endif
			error = ENOTTY;
		}
		if (error == 0 && com == TIOCSCTTY) {

			/* Do nothing if reassigning same control tty */
			sx_slock(&proctree_lock);
			if (td->td_proc->p_session->s_ttyvp == vp) {
				sx_sunlock(&proctree_lock);
				return (0);
			}

			vpold = td->td_proc->p_session->s_ttyvp;
			VREF(vp);
			SESS_LOCK(td->td_proc->p_session);
			td->td_proc->p_session->s_ttyvp = vp;
			SESS_UNLOCK(td->td_proc->p_session);

			sx_sunlock(&proctree_lock);

			/* Get rid of reference to old control tty */
			if (vpold)
				vrele(vpold);
		}
		return (error);
	}
}

/*
 * File table vnode poll routine.
 */
static int
vn_poll(fp, events, active_cred, td)
	struct file *fp;
	int events;
	struct ucred *active_cred;
	struct thread *td;
{
	struct vnode *vp;
#ifdef MAC
	int error;
#endif

	vp = fp->f_data;
#ifdef MAC
	vn_lock(vp, LK_EXCLUSIVE | LK_RETRY, td);
	error = mac_check_vnode_poll(active_cred, fp->f_cred, vp);
	VOP_UNLOCK(vp, 0, td);
	if (error)
		return (error);
#endif

	return (VOP_POLL(vp, events, fp->f_cred, td));
}

/*
 * Check that the vnode is still valid, and if so
 * acquire requested lock.
 */
int
#ifndef	DEBUG_LOCKS
vn_lock(vp, flags, td)
#else
debug_vn_lock(vp, flags, td, filename, line)
#endif
	struct vnode *vp;
	int flags;
	struct thread *td;
#ifdef	DEBUG_LOCKS
	const char *filename;
	int line;
#endif
{
	int error;

	do {
		if ((flags & LK_INTERLOCK) == 0)
			VI_LOCK(vp);
		if ((vp->v_iflag & VI_XLOCK) && vp->v_vxproc != curthread) {
			vp->v_iflag |= VI_XWANT;
			msleep(vp, VI_MTX(vp), PINOD, "vn_lock", 0);
			error = ENOENT;
			if ((flags & LK_RETRY) == 0) {
				VI_UNLOCK(vp);
				return (error);
			}
		} 
#ifdef	DEBUG_LOCKS
		vp->filename = filename;
		vp->line = line;
#endif
		/*
		 * lockmgr drops interlock before it will return for
		 * any reason.  So force the code above to relock it.
		 */
		error = VOP_LOCK(vp, flags | LK_NOPAUSE | LK_INTERLOCK, td);
		flags &= ~LK_INTERLOCK;
	} while (flags & LK_RETRY && error != 0);
	return (error);
}

/*
 * File table vnode close routine.
 */
static int
vn_closefile(fp, td)
	struct file *fp;
	struct thread *td;
{

	fp->f_ops = &badfileops;
	return (vn_close(fp->f_data, fp->f_flag, fp->f_cred, td));
}

/*
 * Preparing to start a filesystem write operation. If the operation is
 * permitted, then we bump the count of operations in progress and
 * proceed. If a suspend request is in progress, we wait until the
 * suspension is over, and then proceed.
 */
int
vn_start_write(vp, mpp, flags)
	struct vnode *vp;
	struct mount **mpp;
	int flags;
{
	struct mount *mp;
	int error;

	/*
	 * If a vnode is provided, get and return the mount point that
	 * to which it will write.
	 */
	if (vp != NULL) {
		if ((error = VOP_GETWRITEMOUNT(vp, mpp)) != 0) {
			*mpp = NULL;
			if (error != EOPNOTSUPP)
				return (error);
			return (0);
		}
	}
	if ((mp = *mpp) == NULL)
		return (0);
	/*
	 * Check on status of suspension.
	 */
	while ((mp->mnt_kern_flag & MNTK_SUSPEND) != 0) {
		if (flags & V_NOWAIT)
			return (EWOULDBLOCK);
		error = tsleep(&mp->mnt_flag, (PUSER - 1) | (flags & PCATCH),
		    "suspfs", 0);
		if (error)
			return (error);
	}
	if (flags & V_XSLEEP)
		return (0);
	mp->mnt_writeopcount++;
	return (0);
}

/*
 * Secondary suspension. Used by operations such as vop_inactive
 * routines that are needed by the higher level functions. These
 * are allowed to proceed until all the higher level functions have
 * completed (indicated by mnt_writeopcount dropping to zero). At that
 * time, these operations are halted until the suspension is over.
 */
int
vn_write_suspend_wait(vp, mp, flags)
	struct vnode *vp;
	struct mount *mp;
	int flags;
{
	int error;

	if (vp != NULL) {
		if ((error = VOP_GETWRITEMOUNT(vp, &mp)) != 0) {
			if (error != EOPNOTSUPP)
				return (error);
			return (0);
		}
	}
	/*
	 * If we are not suspended or have not yet reached suspended
	 * mode, then let the operation proceed.
	 */
	if (mp == NULL || (mp->mnt_kern_flag & MNTK_SUSPENDED) == 0)
		return (0);
	if (flags & V_NOWAIT)
		return (EWOULDBLOCK);
	/*
	 * Wait for the suspension to finish.
	 */
	return (tsleep(&mp->mnt_flag, (PUSER - 1) | (flags & PCATCH),
	    "suspfs", 0));
}

/*
 * Filesystem write operation has completed. If we are suspending and this
 * operation is the last one, notify the suspender that the suspension is
 * now in effect.
 */
void
vn_finished_write(mp)
	struct mount *mp;
{

	if (mp == NULL)
		return;
	mp->mnt_writeopcount--;
	if (mp->mnt_writeopcount < 0)
		panic("vn_finished_write: neg cnt");
	if ((mp->mnt_kern_flag & MNTK_SUSPEND) != 0 &&
	    mp->mnt_writeopcount <= 0)
		wakeup(&mp->mnt_writeopcount);
}

/*
 * Request a filesystem to suspend write operations.
 */
int
vfs_write_suspend(mp)
	struct mount *mp;
{
	struct thread *td = curthread;
	int error;

	if (mp->mnt_kern_flag & MNTK_SUSPEND)
		return (0);
	mp->mnt_kern_flag |= MNTK_SUSPEND;
	if (mp->mnt_writeopcount > 0)
		(void) tsleep(&mp->mnt_writeopcount, PUSER - 1, "suspwt", 0);
	if ((error = VFS_SYNC(mp, MNT_WAIT, td->td_ucred, td)) != 0) {
		vfs_write_resume(mp);
		return (error);
	}
	mp->mnt_kern_flag |= MNTK_SUSPENDED;
	return (0);
}

/*
 * Request a filesystem to resume write operations.
 */
void
vfs_write_resume(mp)
	struct mount *mp;
{

	if ((mp->mnt_kern_flag & MNTK_SUSPEND) == 0)
		return;
	mp->mnt_kern_flag &= ~(MNTK_SUSPEND | MNTK_SUSPENDED);
	wakeup(&mp->mnt_writeopcount);
	wakeup(&mp->mnt_flag);
}

/*
 * Implement kqueues for files by translating it to vnode operation.
 */
static int
vn_kqfilter(struct file *fp, struct knote *kn)
{

	return (VOP_KQFILTER(fp->f_data, kn));
}

/*
 * Simplified in-kernel wrapper calls for extended attribute access.
 * Both calls pass in a NULL credential, authorizing as "kernel" access.
 * Set IO_NODELOCKED in ioflg if the vnode is already locked.
 */
int
vn_extattr_get(struct vnode *vp, int ioflg, int attrnamespace,
    const char *attrname, int *buflen, char *buf, struct thread *td)
{
	struct uio	auio;
	struct iovec	iov;
	int	error;

	iov.iov_len = *buflen;
	iov.iov_base = buf;

	auio.uio_iov = &iov;
	auio.uio_iovcnt = 1;
	auio.uio_rw = UIO_READ;
	auio.uio_segflg = UIO_SYSSPACE;
	auio.uio_td = td;
	auio.uio_offset = 0;
	auio.uio_resid = *buflen;

	if ((ioflg & IO_NODELOCKED) == 0)
		vn_lock(vp, LK_EXCLUSIVE | LK_RETRY, td);

	/* authorize attribute retrieval as kernel */
	error = VOP_GETEXTATTR(vp, attrnamespace, attrname, &auio, NULL, NULL,
	    td);

	if ((ioflg & IO_NODELOCKED) == 0)
		VOP_UNLOCK(vp, 0, td);

	if (error == 0) {
		*buflen = *buflen - auio.uio_resid;
	}

	return (error);
}

/*
 * XXX failure mode if partially written?
 */
int
vn_extattr_set(struct vnode *vp, int ioflg, int attrnamespace,
    const char *attrname, int buflen, char *buf, struct thread *td)
{
	struct uio	auio;
	struct iovec	iov;
	struct mount	*mp;
	int	error;

	iov.iov_len = buflen;
	iov.iov_base = buf;

	auio.uio_iov = &iov;
	auio.uio_iovcnt = 1;
	auio.uio_rw = UIO_WRITE;
	auio.uio_segflg = UIO_SYSSPACE;
	auio.uio_td = td;
	auio.uio_offset = 0;
	auio.uio_resid = buflen;

	if ((ioflg & IO_NODELOCKED) == 0) {
		if ((error = vn_start_write(vp, &mp, V_WAIT)) != 0)
			return (error);
		vn_lock(vp, LK_EXCLUSIVE | LK_RETRY, td);
	}

	/* authorize attribute setting as kernel */
	error = VOP_SETEXTATTR(vp, attrnamespace, attrname, &auio, NULL, td);

	if ((ioflg & IO_NODELOCKED) == 0) {
		vn_finished_write(mp);
		VOP_UNLOCK(vp, 0, td);
	}

	return (error);
}

int
vn_extattr_rm(struct vnode *vp, int ioflg, int attrnamespace,
    const char *attrname, struct thread *td)
{
	struct mount	*mp;
	int	error;

	if ((ioflg & IO_NODELOCKED) == 0) {
		if ((error = vn_start_write(vp, &mp, V_WAIT)) != 0)
			return (error);
		vn_lock(vp, LK_EXCLUSIVE | LK_RETRY, td);
	}

	/* authorize attribute removal as kernel */
	error = VOP_SETEXTATTR(vp, attrnamespace, attrname, NULL, NULL, td);

	if ((ioflg & IO_NODELOCKED) == 0) {
		vn_finished_write(mp);
		VOP_UNLOCK(vp, 0, td);
	}

	return (error);
}
OpenPOWER on IntegriCloud