aboutsummaryrefslogtreecommitdiff
path: root/sys/fs/procfs/procfs_vnops.c
blob: ffab3e8a02fa8cba4091e06c2b7a3afc070af0cb (plain) (blame)
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
/*
 * Copyright (c) 1993, 1995 Jan-Simon Pendry
 * Copyright (c) 1993, 1995
 *	The Regents of the University of California.  All rights reserved.
 *
 * This code is derived from software contributed to Berkeley by
 * Jan-Simon Pendry.
 *
 * 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.
 *
 *	@(#)procfs_vnops.c	8.18 (Berkeley) 5/21/95
 *
 *	$Id: procfs_vnops.c,v 1.64 1999/01/27 22:42:07 dillon Exp $
 */

/*
 * procfs vnode interface
 */

#include <sys/param.h>
#include <sys/systm.h>
#include <sys/time.h>
#include <sys/kernel.h>
#include <sys/lock.h>
#include <sys/fcntl.h>
#include <sys/proc.h>
#include <sys/signalvar.h>
#include <sys/vnode.h>
#include <sys/namei.h>
#include <sys/dirent.h>
#include <machine/reg.h>
#include <vm/vm_zone.h>
#include <miscfs/procfs/procfs.h>
#include <sys/pioctl.h>

static int	procfs_abortop __P((struct vop_abortop_args *));
static int	procfs_access __P((struct vop_access_args *));
static int	procfs_badop __P((void));
static int	procfs_bmap __P((struct vop_bmap_args *));
static int	procfs_close __P((struct vop_close_args *));
static int	procfs_getattr __P((struct vop_getattr_args *));
static int	procfs_inactive __P((struct vop_inactive_args *));
static int	procfs_ioctl __P((struct vop_ioctl_args *));
static int	procfs_lookup __P((struct vop_lookup_args *));
static int	procfs_open __P((struct vop_open_args *));
static int	procfs_print __P((struct vop_print_args *));
static int	procfs_readdir __P((struct vop_readdir_args *));
static int	procfs_readlink __P((struct vop_readlink_args *));
static int	procfs_reclaim __P((struct vop_reclaim_args *));
static int	procfs_setattr __P((struct vop_setattr_args *));

/*
 * This is a list of the valid names in the
 * process-specific sub-directories.  It is
 * used in procfs_lookup and procfs_readdir
 */
static struct proc_target {
	u_char	pt_type;
	u_char	pt_namlen;
	char	*pt_name;
	pfstype	pt_pfstype;
	int	(*pt_valid) __P((struct proc *p));
} proc_targets[] = {
#define N(s) sizeof(s)-1, s
	/*	  name		type		validp */
	{ DT_DIR, N("."),	Pproc,		NULL },
	{ DT_DIR, N(".."),	Proot,		NULL },
	{ DT_REG, N("file"),	Pfile,		procfs_validfile },
	{ DT_REG, N("mem"),	Pmem,		NULL },
	{ DT_REG, N("regs"),	Pregs,		procfs_validregs },
	{ DT_REG, N("fpregs"),	Pfpregs,	procfs_validfpregs },
	{ DT_REG, N("ctl"),	Pctl,		NULL },
	{ DT_REG, N("status"),	Pstatus,	NULL },
	{ DT_REG, N("note"),	Pnote,		NULL },
	{ DT_REG, N("notepg"),	Pnotepg,	NULL },
	{ DT_REG, N("map"), 	Pmap,		procfs_validmap },
	{ DT_REG, N("etype"),	Ptype,		procfs_validtype },
	{ DT_REG, N("cmdline"),	Pcmdline,	NULL },
#undef N
};
static const int nproc_targets = sizeof(proc_targets) / sizeof(proc_targets[0]);

static pid_t atopid __P((const char *, u_int));

/*
 * set things up for doing i/o on
 * the pfsnode (vp).  (vp) is locked
 * on entry, and should be left locked
 * on exit.
 *
 * for procfs we don't need to do anything
 * in particular for i/o.  all that is done
 * is to support exclusive open on process
 * memory images.
 */
static int
procfs_open(ap)
	struct vop_open_args /* {
		struct vnode *a_vp;
		int  a_mode;
		struct ucred *a_cred;
		struct proc *a_p;
	} */ *ap;
{
	struct pfsnode *pfs = VTOPFS(ap->a_vp);
	struct proc *p1, *p2;

	p2 = PFIND(pfs->pfs_pid);
	if (p2 == NULL)
		return (ENOENT);

	switch (pfs->pfs_type) {
	case Pmem:
		if (((pfs->pfs_flags & FWRITE) && (ap->a_mode & O_EXCL)) ||
		    ((pfs->pfs_flags & O_EXCL) && (ap->a_mode & FWRITE)))
			return (EBUSY);

		p1 = ap->a_p;
		if (!CHECKIO(p1, p2) &&
		    !procfs_kmemaccess(p1))
			return (EPERM);

		if (ap->a_mode & FWRITE)
			pfs->pfs_flags = ap->a_mode & (FWRITE|O_EXCL);

		return (0);

	default:
		break;
	}

	return (0);
}

/*
 * close the pfsnode (vp) after doing i/o.
 * (vp) is not locked on entry or exit.
 *
 * nothing to do for procfs other than undo
 * any exclusive open flag (see _open above).
 */
static int
procfs_close(ap)
	struct vop_close_args /* {
		struct vnode *a_vp;
		int  a_fflag;
		struct ucred *a_cred;
		struct proc *a_p;
	} */ *ap;
{
	struct pfsnode *pfs = VTOPFS(ap->a_vp);
	struct proc *p;

	switch (pfs->pfs_type) {
	case Pmem:
		if ((ap->a_fflag & FWRITE) && (pfs->pfs_flags & O_EXCL))
			pfs->pfs_flags &= ~(FWRITE|O_EXCL);
		/*
		 * This rather complicated-looking code is trying to
		 * determine if this was the last close on this particular
		 * vnode.  While one would expect v_usecount to be 1 at
		 * that point, it seems that (according to John Dyson)
		 * the VM system will bump up the usecount.  So:  if the
		 * usecount is 2, and VOBJBUF is set, then this is really
		 * the last close.  Otherwise, if the usecount is < 2
		 * then it is definitely the last close.
		 * If this is the last close, then it checks to see if
		 * the target process has PF_LINGER set in p_pfsflags,
		 * if this is *not* the case, then the process' stop flags
		 * are cleared, and the process is woken up.  This is
		 * to help prevent the case where a process has been
		 * told to stop on an event, but then the requesting process
		 * has gone away or forgotten about it.
		 */
		if ((ap->a_vp->v_usecount < 2)
		    && (p = pfind(pfs->pfs_pid))
		    && !(p->p_pfsflags & PF_LINGER)) {
			p->p_stops = 0;
			p->p_step = 0;
			wakeup(&p->p_step);
		}
		break;
	default:
		break;
	}

	return (0);
}

/*
 * do an ioctl operation on a pfsnode (vp).
 * (vp) is not locked on entry or exit.
 */
static int
procfs_ioctl(ap)
	struct vop_ioctl_args *ap;
{
	struct pfsnode *pfs = VTOPFS(ap->a_vp);
	struct proc *procp, *p;
	int error;
	int signo;
	struct procfs_status *psp;
	unsigned char flags;

	p = ap->a_p;
	procp = pfind(pfs->pfs_pid);
	if (procp == NULL) {
		return ENOTTY;
	}

	if (!CHECKIO(p, procp))
		return EPERM;

	switch (ap->a_command) {
	case PIOCBIS:
	  procp->p_stops |= *(unsigned int*)ap->a_data;
	  break;
	case PIOCBIC:
	  procp->p_stops &= ~*(unsigned int*)ap->a_data;
	  break;
	case PIOCSFL:
	  /*
	   * NFLAGS is "non-suser_xxx flags" -- currently, only
	   * PFS_ISUGID ("ignore set u/g id");
	   */
#define NFLAGS	(PF_ISUGID)
	  flags = (unsigned char)*(unsigned int*)ap->a_data;
	  if (flags & NFLAGS && (error = suser(p)))
	    return error;
	  procp->p_pfsflags = flags;
	  break;
	case PIOCGFL:
	  *(unsigned int*)ap->a_data = (unsigned int)procp->p_pfsflags;
	case PIOCSTATUS:
	  psp = (struct procfs_status *)ap->a_data;
	  psp->state = (procp->p_step == 0);
	  psp->flags = procp->p_pfsflags;
	  psp->events = procp->p_stops;
	  if (procp->p_step) {
	    psp->why = procp->p_stype;
	    psp->val = procp->p_xstat;
	  } else {
	    psp->why = psp->val = 0;	/* Not defined values */
	  }
	  break;
	case PIOCWAIT:
	  psp = (struct procfs_status *)ap->a_data;
	  if (procp->p_step == 0) {
	    error = tsleep(&procp->p_stype, PWAIT | PCATCH, "piocwait", 0);
	    if (error)
	      return error;
	  }
	  psp->state = 1;	/* It stopped */
	  psp->flags = procp->p_pfsflags;
	  psp->events = procp->p_stops;
	  psp->why = procp->p_stype;	/* why it stopped */
	  psp->val = procp->p_xstat;	/* any extra info */
	  break;
	case PIOCCONT:	/* Restart a proc */
	  if (procp->p_step == 0)
	    return EINVAL;	/* Can only start a stopped process */
	  if ((signo = *(int*)ap->a_data) != 0) {
	    if (signo >= NSIG || signo <= 0)
	      return EINVAL;
	    psignal(procp, signo);
	  }
	  procp->p_step = 0;
	  wakeup(&procp->p_step);
	  break;
	default:
	  return (ENOTTY);
	}
	return 0;
}

/*
 * do block mapping for pfsnode (vp).
 * since we don't use the buffer cache
 * for procfs this function should never
 * be called.  in any case, it's not clear
 * what part of the kernel ever makes use
 * of this function.  for sanity, this is the
 * usual no-op bmap, although returning
 * (EIO) would be a reasonable alternative.
 */
static int
procfs_bmap(ap)
	struct vop_bmap_args /* {
		struct vnode *a_vp;
		daddr_t  a_bn;
		struct vnode **a_vpp;
		daddr_t *a_bnp;
		int *a_runp;
	} */ *ap;
{

	if (ap->a_vpp != NULL)
		*ap->a_vpp = ap->a_vp;
	if (ap->a_bnp != NULL)
		*ap->a_bnp = ap->a_bn;
	if (ap->a_runp != NULL)
		*ap->a_runp = 0;
	return (0);
}

/*
 * procfs_inactive is called when the pfsnode
 * is vrele'd and the reference count goes
 * to zero.  (vp) will be on the vnode free
 * list, so to get it back vget() must be
 * used.
 *
 * (vp) is locked on entry, but must be unlocked on exit.
 */
static int
procfs_inactive(ap)
	struct vop_inactive_args /* {
		struct vnode *a_vp;
	} */ *ap;
{
	struct vnode *vp = ap->a_vp;

	VOP_UNLOCK(vp, 0, ap->a_p);

	return (0);
}

/*
 * _reclaim is called when getnewvnode()
 * wants to make use of an entry on the vnode
 * free list.  at this time the filesystem needs
 * to free any private data and remove the node
 * from any private lists.
 */
static int
procfs_reclaim(ap)
	struct vop_reclaim_args /* {
		struct vnode *a_vp;
	} */ *ap;
{

	return (procfs_freevp(ap->a_vp));
}

/*
 * _print is used for debugging.
 * just print a readable description
 * of (vp).
 */
static int
procfs_print(ap)
	struct vop_print_args /* {
		struct vnode *a_vp;
	} */ *ap;
{
	struct pfsnode *pfs = VTOPFS(ap->a_vp);

	printf("tag VT_PROCFS, type %d, pid %ld, mode %x, flags %lx\n",
	    pfs->pfs_type, (long)pfs->pfs_pid, pfs->pfs_mode, pfs->pfs_flags);
	return (0);
}

/*
 * _abortop is called when operations such as
 * rename and create fail.  this entry is responsible
 * for undoing any side-effects caused by the lookup.
 * this will always include freeing the pathname buffer.
 */
static int
procfs_abortop(ap)
	struct vop_abortop_args /* {
		struct vnode *a_dvp;
		struct componentname *a_cnp;
	} */ *ap;
{

	if ((ap->a_cnp->cn_flags & (HASBUF | SAVESTART)) == HASBUF)
		zfree(namei_zone, ap->a_cnp->cn_pnbuf);
	return (0);
}

/*
 * generic entry point for unsupported operations
 */
static int
procfs_badop()
{

	return (EIO);
}

/*
 * Invent attributes for pfsnode (vp) and store
 * them in (vap).
 * Directories lengths are returned as zero since
 * any real length would require the genuine size
 * to be computed, and nothing cares anyway.
 *
 * this is relatively minimal for procfs.
 */
static int
procfs_getattr(ap)
	struct vop_getattr_args /* {
		struct vnode *a_vp;
		struct vattr *a_vap;
		struct ucred *a_cred;
		struct proc *a_p;
	} */ *ap;
{
	struct pfsnode *pfs = VTOPFS(ap->a_vp);
	struct vattr *vap = ap->a_vap;
	struct proc *procp;
	int error;

	/*
	 * First make sure that the process and its credentials 
	 * still exist.
	 */
	switch (pfs->pfs_type) {
	case Proot:
	case Pcurproc:
		procp = 0;
		break;

	default:
		procp = PFIND(pfs->pfs_pid);
		if (procp == 0 || procp->p_cred == NULL ||
		    procp->p_ucred == NULL)
			return (ENOENT);
	}

	error = 0;

	/* start by zeroing out the attributes */
	VATTR_NULL(vap);

	/* next do all the common fields */
	vap->va_type = ap->a_vp->v_type;
	vap->va_mode = pfs->pfs_mode;
	vap->va_fileid = pfs->pfs_fileno;
	vap->va_flags = 0;
	vap->va_blocksize = PAGE_SIZE;
	vap->va_bytes = vap->va_size = 0;

	/*
	 * Make all times be current TOD.
	 * It would be possible to get the process start
	 * time from the p_stat structure, but there's
	 * no "file creation" time stamp anyway, and the
	 * p_stat structure is not addressible if u. gets
	 * swapped out for that process.
	 */
	nanotime(&vap->va_ctime);
	vap->va_atime = vap->va_mtime = vap->va_ctime;

	/*
	 * If the process has exercised some setuid or setgid
	 * privilege, then rip away read/write permission so
	 * that only root can gain access.
	 */
	switch (pfs->pfs_type) {
	case Pctl:
	case Pregs:
	case Pfpregs:
		if (procp->p_flag & P_SUGID)
			vap->va_mode &= ~((VREAD|VWRITE)|
					  ((VREAD|VWRITE)>>3)|
					  ((VREAD|VWRITE)>>6));
		break;
	case Pmem:
		/* Retain group kmem readablity. */
		if (procp->p_flag & P_SUGID)
			vap->va_mode &= ~(VREAD|VWRITE);
		break;
	default:
		break;
	}

	/*
	 * now do the object specific fields
	 *
	 * The size could be set from struct reg, but it's hardly
	 * worth the trouble, and it puts some (potentially) machine
	 * dependent data into this machine-independent code.  If it
	 * becomes important then this function should break out into
	 * a per-file stat function in the corresponding .c file.
	 */

	switch (pfs->pfs_type) {
	case Proot:
		/*
		 * Set nlink to 1 to tell fts(3) we don't actually know.
		 */
		vap->va_nlink = 1;
		vap->va_uid = 0;
		vap->va_gid = 0;
		vap->va_size = vap->va_bytes = DEV_BSIZE;
		break;

	case Pcurproc: {
		char buf[16];		/* should be enough */
		vap->va_nlink = 1;
		vap->va_uid = 0;
		vap->va_gid = 0;
		vap->va_size = vap->va_bytes =
		    snprintf(buf, sizeof(buf), "%ld", (long)curproc->p_pid);
		break;
	}

	case Pproc:
		vap->va_nlink = nproc_targets;
		vap->va_uid = procp->p_ucred->cr_uid;
		vap->va_gid = procp->p_ucred->cr_gid;
		vap->va_size = vap->va_bytes = DEV_BSIZE;
		break;

	case Pfile:
		error = EOPNOTSUPP;
		break;

	case Pmem:
		vap->va_nlink = 1;
		/*
		 * If we denied owner access earlier, then we have to
		 * change the owner to root - otherwise 'ps' and friends
		 * will break even though they are setgid kmem. *SIGH*
		 */
		if (procp->p_flag & P_SUGID)
			vap->va_uid = 0;
		else
			vap->va_uid = procp->p_ucred->cr_uid;
		vap->va_gid = KMEM_GROUP;
		break;

	case Ptype:
	case Pmap:
	case Pregs:
		vap->va_bytes = vap->va_size = sizeof(struct reg);
		vap->va_nlink = 1;
		vap->va_uid = procp->p_ucred->cr_uid;
		vap->va_gid = procp->p_ucred->cr_gid;
		break;

	case Pfpregs:
		vap->va_bytes = vap->va_size = sizeof(struct fpreg);

	case Pctl:
	case Pstatus:
	case Pnote:
	case Pnotepg:
	case Pcmdline:
		vap->va_nlink = 1;
		vap->va_uid = procp->p_ucred->cr_uid;
		vap->va_gid = procp->p_ucred->cr_gid;
		break;

	default:
		panic("procfs_getattr");
	}

	return (error);
}

static int
procfs_setattr(ap)
	struct vop_setattr_args /* {
		struct vnode *a_vp;
		struct vattr *a_vap;
		struct ucred *a_cred;
		struct proc *a_p;
	} */ *ap;
{

	if (ap->a_vap->va_flags != VNOVAL)
		return (EOPNOTSUPP);

	/*
	 * just fake out attribute setting
	 * it's not good to generate an error
	 * return, otherwise things like creat()
	 * will fail when they try to set the
	 * file length to 0.  worse, this means
	 * that echo $note > /proc/$pid/note will fail.
	 */

	return (0);
}

/*
 * implement access checking.
 *
 * something very similar to this code is duplicated
 * throughout the 4bsd kernel and should be moved
 * into kern/vfs_subr.c sometime.
 *
 * actually, the check for super-user is slightly
 * broken since it will allow read access to write-only
 * objects.  this doesn't cause any particular trouble
 * but does mean that the i/o entry points need to check
 * that the operation really does make sense.
 */
static int
procfs_access(ap)
	struct vop_access_args /* {
		struct vnode *a_vp;
		int a_mode;
		struct ucred *a_cred;
		struct proc *a_p;
	} */ *ap;
{
	struct vattr *vap;
	struct vattr vattr;
	int error;

	/*
	 * If you're the super-user,
	 * you always get access.
	 */
	if (ap->a_cred->cr_uid == 0)
		return (0);

	vap = &vattr;
	error = VOP_GETATTR(ap->a_vp, vap, ap->a_cred, ap->a_p);
	if (error)
		return (error);

	/*
	 * Access check is based on only one of owner, group, public.
	 * If not owner, then check group. If not a member of the
	 * group, then check public access.
	 */
	if (ap->a_cred->cr_uid != vap->va_uid) {
		gid_t *gp;
		int i;

		ap->a_mode >>= 3;
		gp = ap->a_cred->cr_groups;
		for (i = 0; i < ap->a_cred->cr_ngroups; i++, gp++)
			if (vap->va_gid == *gp)
				goto found;
		ap->a_mode >>= 3;
found:
		;
	}

	if ((vap->va_mode & ap->a_mode) == ap->a_mode)
		return (0);

	return (EACCES);
}

/*
 * lookup.  this is incredibly complicated in the
 * general case, however for most pseudo-filesystems
 * very little needs to be done.
 *
 * unless you want to get a migraine, just make sure your
 * filesystem doesn't do any locking of its own.  otherwise
 * read and inwardly digest ufs_lookup().
 */
static int
procfs_lookup(ap)
	struct vop_lookup_args /* {
		struct vnode * a_dvp;
		struct vnode ** a_vpp;
		struct componentname * a_cnp;
	} */ *ap;
{
	struct componentname *cnp = ap->a_cnp;
	struct vnode **vpp = ap->a_vpp;
	struct vnode *dvp = ap->a_dvp;
	char *pname = cnp->cn_nameptr;
	struct proc *curp = cnp->cn_proc;
	struct proc_target *pt;
	struct vnode *fvp;
	pid_t pid;
	struct pfsnode *pfs;
	struct proc *p;
	int i;

	*vpp = NULL;

	if (cnp->cn_nameiop == DELETE || cnp->cn_nameiop == RENAME)
		return (EROFS);

	if (cnp->cn_namelen == 1 && *pname == '.') {
		*vpp = dvp;
		VREF(dvp);
		/* vn_lock(dvp, LK_EXCLUSIVE | LK_RETRY, curp); */
		return (0);
	}

	pfs = VTOPFS(dvp);
	switch (pfs->pfs_type) {
	case Proot:
		if (cnp->cn_flags & ISDOTDOT)
			return (EIO);

		if (CNEQ(cnp, "curproc", 7))
			return (procfs_allocvp(dvp->v_mount, vpp, 0, Pcurproc));

		pid = atopid(pname, cnp->cn_namelen);
		if (pid == NO_PID)
			break;

		p = PFIND(pid);
		if (p == 0)
			break;

		return (procfs_allocvp(dvp->v_mount, vpp, pid, Pproc));

	case Pproc:
		if (cnp->cn_flags & ISDOTDOT)
			return (procfs_root(dvp->v_mount, vpp));

		p = PFIND(pfs->pfs_pid);
		if (p == 0)
			break;

		for (pt = proc_targets, i = 0; i < nproc_targets; pt++, i++) {
			if (cnp->cn_namelen == pt->pt_namlen &&
			    bcmp(pt->pt_name, pname, cnp->cn_namelen) == 0 &&
			    (pt->pt_valid == NULL || (*pt->pt_valid)(p)))
				goto found;
		}
		break;

	found:
		if (pt->pt_pfstype == Pfile) {
			fvp = procfs_findtextvp(p);
			/* We already checked that it exists. */
			VREF(fvp);
			vn_lock(fvp, LK_EXCLUSIVE | LK_RETRY, curp);
			*vpp = fvp;
			return (0);
		}

		return (procfs_allocvp(dvp->v_mount, vpp, pfs->pfs_pid,
		    pt->pt_pfstype));

	default:
		return (ENOTDIR);
	}

	return (cnp->cn_nameiop == LOOKUP ? ENOENT : EROFS);
}

/*
 * Does this process have a text file?
 */
int
procfs_validfile(p)
	struct proc *p;
{

	return (procfs_findtextvp(p) != NULLVP);
}

/*
 * readdir returns directory entries from pfsnode (vp).
 *
 * the strategy here with procfs is to generate a single
 * directory entry at a time (struct pfsdent) and then
 * copy that out to userland using uiomove.  a more efficent
 * though more complex implementation, would try to minimize
 * the number of calls to uiomove().  for procfs, this is
 * hardly worth the added code complexity.
 *
 * this should just be done through read()
 */
static int
procfs_readdir(ap)
	struct vop_readdir_args /* {
		struct vnode *a_vp;
		struct uio *a_uio;
		struct ucred *a_cred;
		int *a_eofflag;
		int *a_ncookies;
		u_long **a_cookies;
	} */ *ap;
{
	struct uio *uio = ap->a_uio;
	struct pfsdent d;
	struct pfsdent *dp = &d;
	struct pfsnode *pfs;
	int count, error, i, off;

	pfs = VTOPFS(ap->a_vp);

	off = (int)uio->uio_offset;
	if (off != uio->uio_offset || off < 0 || (u_int)off % UIO_MX != 0 ||
	    uio->uio_resid < UIO_MX)
		return (EINVAL);

	error = 0;
	count = 0;
	i = (u_int)off / UIO_MX;

	switch (pfs->pfs_type) {
	/*
	 * this is for the process-specific sub-directories.
	 * all that is needed to is copy out all the entries
	 * from the procent[] table (top of this file).
	 */
	case Pproc: {
		struct proc *p;
		struct proc_target *pt;

		p = PFIND(pfs->pfs_pid);
		if (p == NULL)
			break;

		for (pt = &proc_targets[i];
		     uio->uio_resid >= UIO_MX && i < nproc_targets; pt++, i++) {
			if (pt->pt_valid && (*pt->pt_valid)(p) == 0)
				continue;

			dp->d_reclen = UIO_MX;
			dp->d_fileno = PROCFS_FILENO(pfs->pfs_pid, pt->pt_pfstype);
			dp->d_namlen = pt->pt_namlen;
			bcopy(pt->pt_name, dp->d_name, pt->pt_namlen + 1);
			dp->d_type = pt->pt_type;

			if ((error = uiomove((caddr_t)dp, UIO_MX, uio)) != 0)
				break;
		}

	    	break;
	    }

	/*
	 * this is for the root of the procfs filesystem
	 * what is needed is a special entry for "curproc"
	 * followed by an entry for each process on allproc
#ifdef PROCFS_ZOMBIE
	 * and zombproc.
#endif
	 */

	case Proot: {
#ifdef PROCFS_ZOMBIE
		int doingzomb = 0;
#endif
		int pcnt = 0;
		volatile struct proc *p = allproc.lh_first;

		for (; p && uio->uio_resid >= UIO_MX; i++, pcnt++) {
			bzero((char *) dp, UIO_MX);
			dp->d_reclen = UIO_MX;

			switch (i) {
			case 0:		/* `.' */
			case 1:		/* `..' */
				dp->d_fileno = PROCFS_FILENO(0, Proot);
				dp->d_namlen = i + 1;
				bcopy("..", dp->d_name, dp->d_namlen);
				dp->d_name[i + 1] = '\0';
				dp->d_type = DT_DIR;
				break;

			case 2:
				dp->d_fileno = PROCFS_FILENO(0, Pcurproc);
				dp->d_namlen = 7;
				bcopy("curproc", dp->d_name, 8);
				dp->d_type = DT_LNK;
				break;

			default:
				while (pcnt < i) {
					pcnt++;
					p = p->p_list.le_next;
					if (!p)
						goto done;
				}
				dp->d_fileno = PROCFS_FILENO(p->p_pid, Pproc);
				dp->d_namlen = sprintf(dp->d_name, "%ld",
				    (long)p->p_pid);
				dp->d_type = DT_REG;
				p = p->p_list.le_next;
				break;
			}

			if ((error = uiomove((caddr_t)dp, UIO_MX, uio)) != 0)
				break;
		}
	done:

#ifdef PROCFS_ZOMBIE
		if (p == 0 && doingzomb == 0) {
			doingzomb = 1;
			p = zombproc.lh_first;
			goto again;
		}
#endif

		break;

	    }

	default:
		error = ENOTDIR;
		break;
	}

	uio->uio_offset = i * UIO_MX;

	return (error);
}

/*
 * readlink reads the link of `curproc'
 */
static int
procfs_readlink(ap)
	struct vop_readlink_args *ap;
{
	char buf[16];		/* should be enough */
	int len;

	if (VTOPFS(ap->a_vp)->pfs_fileno != PROCFS_FILENO(0, Pcurproc))
		return (EINVAL);

	len = snprintf(buf, sizeof(buf), "%ld", (long)curproc->p_pid);

	return (uiomove((caddr_t)buf, len, ap->a_uio));
}

/*
 * convert decimal ascii to pid_t
 */
static pid_t
atopid(b, len)
	const char *b;
	u_int len;
{
	pid_t p = 0;

	while (len--) {
		char c = *b++;
		if (c < '0' || c > '9')
			return (NO_PID);
		p = 10 * p + (c - '0');
		if (p > PID_MAX)
			return (NO_PID);
	}

	return (p);
}

/*
 * procfs vnode operations.
 */
vop_t **procfs_vnodeop_p;
static struct vnodeopv_entry_desc procfs_vnodeop_entries[] = {
	{ &vop_default_desc,		(vop_t *) vop_defaultop },
	{ &vop_abortop_desc,		(vop_t *) procfs_abortop },
	{ &vop_access_desc,		(vop_t *) procfs_access },
	{ &vop_advlock_desc,		(vop_t *) procfs_badop },
	{ &vop_bmap_desc,		(vop_t *) procfs_bmap },
	{ &vop_close_desc,		(vop_t *) procfs_close },
	{ &vop_create_desc,		(vop_t *) procfs_badop },
	{ &vop_getattr_desc,		(vop_t *) procfs_getattr },
	{ &vop_inactive_desc,		(vop_t *) procfs_inactive },
	{ &vop_link_desc,		(vop_t *) procfs_badop },
	{ &vop_lookup_desc,		(vop_t *) procfs_lookup },
	{ &vop_mkdir_desc,		(vop_t *) procfs_badop },
	{ &vop_mknod_desc,		(vop_t *) procfs_badop },
	{ &vop_open_desc,		(vop_t *) procfs_open },
	{ &vop_pathconf_desc,		(vop_t *) vop_stdpathconf },
	{ &vop_print_desc,		(vop_t *) procfs_print },
	{ &vop_read_desc,		(vop_t *) procfs_rw },
	{ &vop_readdir_desc,		(vop_t *) procfs_readdir },
	{ &vop_readlink_desc,		(vop_t *) procfs_readlink },
	{ &vop_reclaim_desc,		(vop_t *) procfs_reclaim },
	{ &vop_remove_desc,		(vop_t *) procfs_badop },
	{ &vop_rename_desc,		(vop_t *) procfs_badop },
	{ &vop_rmdir_desc,		(vop_t *) procfs_badop },
	{ &vop_setattr_desc,		(vop_t *) procfs_setattr },
	{ &vop_symlink_desc,		(vop_t *) procfs_badop },
	{ &vop_write_desc,		(vop_t *) procfs_rw },
	{ &vop_ioctl_desc,		(vop_t *) procfs_ioctl },
	{ NULL, NULL }
};
static struct vnodeopv_desc procfs_vnodeop_opv_desc =
	{ &procfs_vnodeop_p, procfs_vnodeop_entries };

VNODEOP_SET(procfs_vnodeop_opv_desc);