summaryrefslogtreecommitdiffstats
path: root/sys/kern/sysv_shm.c
blob: 80fa0de04bb206ffd0eb6e04c944e8a0cb418d9f (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
/*	$NetBSD: sysv_shm.c,v 1.23 1994/07/04 23:25:12 glass Exp $	*/
/*
 * Copyright (c) 1994 Adam Glass and Charles Hannum.  All rights reserved.
 *
 * 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 Adam Glass and Charles
 *	Hannum.
 * 4. The names of the authors may not be used to endorse or promote products
 *    derived from this software without specific prior written permission.
 *
 * THIS SOFTWARE IS PROVIDED BY THE AUTHORS ``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 AUTHORS 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.
 */

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

#include "opt_compat.h"
#include "opt_sysvipc.h"

#include <sys/param.h>
#include <sys/systm.h>
#include <sys/kernel.h>
#include <sys/lock.h>
#include <sys/sysctl.h>
#include <sys/shm.h>
#include <sys/proc.h>
#include <sys/malloc.h>
#include <sys/mman.h>
#include <sys/mutex.h>
#include <sys/stat.h>
#include <sys/syscall.h>
#include <sys/syscallsubr.h>
#include <sys/sysent.h>
#include <sys/sysproto.h>
#include <sys/jail.h>

#include <vm/vm.h>
#include <vm/vm_param.h>
#include <vm/pmap.h>
#include <vm/vm_object.h>
#include <vm/vm_map.h>
#include <vm/vm_page.h>
#include <vm/vm_pager.h>

static MALLOC_DEFINE(M_SHM, "shm", "SVID compatible shared memory segments");

struct oshmctl_args;
static int oshmctl(struct thread *td, struct oshmctl_args *uap);

static int shmget_allocate_segment(struct thread *td,
    struct shmget_args *uap, int mode);
static int shmget_existing(struct thread *td, struct shmget_args *uap,
    int mode, int segnum);

/* XXX casting to (sy_call_t *) is bogus, as usual. */
static sy_call_t *shmcalls[] = {
	(sy_call_t *)shmat, (sy_call_t *)oshmctl,
	(sy_call_t *)shmdt, (sy_call_t *)shmget,
	(sy_call_t *)shmctl
};

#define	SHMSEG_FREE     	0x0200
#define	SHMSEG_REMOVED  	0x0400
#define	SHMSEG_ALLOCATED	0x0800
#define	SHMSEG_WANTED		0x1000

static int shm_last_free, shm_nused, shm_committed, shmalloced;
static struct shmid_ds	*shmsegs;

struct shm_handle {
	/* vm_offset_t kva; */
	vm_object_t shm_object;
};

struct shmmap_state {
	vm_offset_t va;
	int shmid;
};

static void shm_deallocate_segment(struct shmid_ds *);
static int shm_find_segment_by_key(key_t);
static struct shmid_ds *shm_find_segment_by_shmid(int);
static struct shmid_ds *shm_find_segment_by_shmidx(int);
static int shm_delete_mapping(struct vmspace *vm, struct shmmap_state *);
static void shmrealloc(void);
static void shminit(void);
static int sysvshm_modload(struct module *, int, void *);
static int shmunload(void);
static void shmexit_myhook(struct vmspace *vm);
static void shmfork_myhook(struct proc *p1, struct proc *p2);
static int sysctl_shmsegs(SYSCTL_HANDLER_ARGS);

/*
 * Tuneable values.
 */
#ifndef SHMMAXPGS
#define	SHMMAXPGS	8192	/* Note: sysv shared memory is swap backed. */
#endif
#ifndef SHMMAX
#define	SHMMAX	(SHMMAXPGS*PAGE_SIZE)
#endif
#ifndef SHMMIN
#define	SHMMIN	1
#endif
#ifndef SHMMNI
#define	SHMMNI	192
#endif
#ifndef SHMSEG
#define	SHMSEG	128
#endif
#ifndef SHMALL
#define	SHMALL	(SHMMAXPGS)
#endif

struct	shminfo shminfo = {
	SHMMAX,
	SHMMIN,
	SHMMNI,
	SHMSEG,
	SHMALL
};

static int shm_use_phys;
static int shm_allow_removed;

SYSCTL_DECL(_kern_ipc);
SYSCTL_INT(_kern_ipc, OID_AUTO, shmmax, CTLFLAG_RW, &shminfo.shmmax, 0, "");
SYSCTL_INT(_kern_ipc, OID_AUTO, shmmin, CTLFLAG_RW, &shminfo.shmmin, 0, "");
SYSCTL_INT(_kern_ipc, OID_AUTO, shmmni, CTLFLAG_RDTUN, &shminfo.shmmni, 0, "");
SYSCTL_INT(_kern_ipc, OID_AUTO, shmseg, CTLFLAG_RDTUN, &shminfo.shmseg, 0, "");
SYSCTL_INT(_kern_ipc, OID_AUTO, shmall, CTLFLAG_RW, &shminfo.shmall, 0, "");
SYSCTL_INT(_kern_ipc, OID_AUTO, shm_use_phys, CTLFLAG_RW,
    &shm_use_phys, 0, "");
SYSCTL_INT(_kern_ipc, OID_AUTO, shm_allow_removed, CTLFLAG_RW,
    &shm_allow_removed, 0, "");
SYSCTL_PROC(_kern_ipc, OID_AUTO, shmsegs, CTLFLAG_RD,
    NULL, 0, sysctl_shmsegs, "", "");

static int
shm_find_segment_by_key(key)
	key_t key;
{
	int i;

	for (i = 0; i < shmalloced; i++)
		if ((shmsegs[i].shm_perm.mode & SHMSEG_ALLOCATED) &&
		    shmsegs[i].shm_perm.key == key)
			return (i);
	return (-1);
}

static struct shmid_ds *
shm_find_segment_by_shmid(int shmid)
{
	int segnum;
	struct shmid_ds *shmseg;

	segnum = IPCID_TO_IX(shmid);
	if (segnum < 0 || segnum >= shmalloced)
		return (NULL);
	shmseg = &shmsegs[segnum];
	if ((shmseg->shm_perm.mode & SHMSEG_ALLOCATED) == 0 ||
	    (!shm_allow_removed &&
	     (shmseg->shm_perm.mode & SHMSEG_REMOVED) != 0) ||
	    shmseg->shm_perm.seq != IPCID_TO_SEQ(shmid))
		return (NULL);
	return (shmseg);
}

static struct shmid_ds *
shm_find_segment_by_shmidx(int segnum)
{
	struct shmid_ds *shmseg;

	if (segnum < 0 || segnum >= shmalloced)
		return (NULL);
	shmseg = &shmsegs[segnum];
	if ((shmseg->shm_perm.mode & SHMSEG_ALLOCATED) == 0 ||
	    (!shm_allow_removed &&
	     (shmseg->shm_perm.mode & SHMSEG_REMOVED) != 0))
		return (NULL);
	return (shmseg);
}

static void
shm_deallocate_segment(shmseg)
	struct shmid_ds *shmseg;
{
	struct shm_handle *shm_handle;
	size_t size;

	GIANT_REQUIRED;

	shm_handle = shmseg->shm_internal;
	vm_object_deallocate(shm_handle->shm_object);
	free(shm_handle, M_SHM);
	shmseg->shm_internal = NULL;
	size = round_page(shmseg->shm_segsz);
	shm_committed -= btoc(size);
	shm_nused--;
	shmseg->shm_perm.mode = SHMSEG_FREE;
}

static int
shm_delete_mapping(struct vmspace *vm, struct shmmap_state *shmmap_s)
{
	struct shmid_ds *shmseg;
	int segnum, result;
	size_t size;

	GIANT_REQUIRED;

	segnum = IPCID_TO_IX(shmmap_s->shmid);
	shmseg = &shmsegs[segnum];
	size = round_page(shmseg->shm_segsz);
	result = vm_map_remove(&vm->vm_map, shmmap_s->va, shmmap_s->va + size);
	if (result != KERN_SUCCESS)
		return (EINVAL);
	shmmap_s->shmid = -1;
	shmseg->shm_dtime = time_second;
	if ((--shmseg->shm_nattch <= 0) &&
	    (shmseg->shm_perm.mode & SHMSEG_REMOVED)) {
		shm_deallocate_segment(shmseg);
		shm_last_free = segnum;
	}
	return (0);
}

#ifndef _SYS_SYSPROTO_H_
struct shmdt_args {
	const void *shmaddr;
};
#endif

/*
 * MPSAFE
 */
int
shmdt(td, uap)
	struct thread *td;
	struct shmdt_args *uap;
{
	struct proc *p = td->td_proc;
	struct shmmap_state *shmmap_s;
	int i;
	int error = 0;

	if (!jail_sysvipc_allowed && jailed(td->td_ucred))
		return (ENOSYS);
	mtx_lock(&Giant);
	shmmap_s = p->p_vmspace->vm_shm;
 	if (shmmap_s == NULL) {
		error = EINVAL;
		goto done2;
	}
	for (i = 0; i < shminfo.shmseg; i++, shmmap_s++) {
		if (shmmap_s->shmid != -1 &&
		    shmmap_s->va == (vm_offset_t)uap->shmaddr) {
			break;
		}
	}
	if (i == shminfo.shmseg) {
		error = EINVAL;
		goto done2;
	}
	error = shm_delete_mapping(p->p_vmspace, shmmap_s);
done2:
	mtx_unlock(&Giant);
	return (error);
}

#ifndef _SYS_SYSPROTO_H_
struct shmat_args {
	int shmid;
	const void *shmaddr;
	int shmflg;
};
#endif

/*
 * MPSAFE
 */
int
kern_shmat(td, shmid, shmaddr, shmflg)
	struct thread *td;
	int shmid;
	const void *shmaddr;
	int shmflg;
{
	struct proc *p = td->td_proc;
	int i, flags;
	struct shmid_ds *shmseg;
	struct shmmap_state *shmmap_s = NULL;
	struct shm_handle *shm_handle;
	vm_offset_t attach_va;
	vm_prot_t prot;
	vm_size_t size;
	int rv;
	int error = 0;

	if (!jail_sysvipc_allowed && jailed(td->td_ucred))
		return (ENOSYS);
	mtx_lock(&Giant);
	shmmap_s = p->p_vmspace->vm_shm;
	if (shmmap_s == NULL) {
		size = shminfo.shmseg * sizeof(struct shmmap_state);
		shmmap_s = malloc(size, M_SHM, M_WAITOK);
		for (i = 0; i < shminfo.shmseg; i++)
			shmmap_s[i].shmid = -1;
		p->p_vmspace->vm_shm = shmmap_s;
	}
	shmseg = shm_find_segment_by_shmid(shmid);
	if (shmseg == NULL) {
		error = EINVAL;
		goto done2;
	}
	error = ipcperm(td, &shmseg->shm_perm,
	    (shmflg & SHM_RDONLY) ? IPC_R : IPC_R|IPC_W);
	if (error)
		goto done2;
	for (i = 0; i < shminfo.shmseg; i++) {
		if (shmmap_s->shmid == -1)
			break;
		shmmap_s++;
	}
	if (i >= shminfo.shmseg) {
		error = EMFILE;
		goto done2;
	}
	size = round_page(shmseg->shm_segsz);
#ifdef VM_PROT_READ_IS_EXEC
	prot = VM_PROT_READ | VM_PROT_EXECUTE;
#else
	prot = VM_PROT_READ;
#endif
	if ((shmflg & SHM_RDONLY) == 0)
		prot |= VM_PROT_WRITE;
	flags = MAP_ANON | MAP_SHARED;
	if (shmaddr) {
		flags |= MAP_FIXED;
		if (shmflg & SHM_RND) {
			attach_va = (vm_offset_t)shmaddr & ~(SHMLBA-1);
		} else if (((vm_offset_t)shmaddr & (SHMLBA-1)) == 0) {
			attach_va = (vm_offset_t)shmaddr;
		} else {
			error = EINVAL;
			goto done2;
		}
	} else {
		/*
		 * This is just a hint to vm_map_find() about where to
		 * put it.
		 */
		attach_va = round_page((vm_offset_t)p->p_vmspace->vm_taddr
		    + maxtsiz + maxdsiz);
	}

	shm_handle = shmseg->shm_internal;
	vm_object_reference(shm_handle->shm_object);
	rv = vm_map_find(&p->p_vmspace->vm_map, shm_handle->shm_object,
		0, &attach_va, size, (flags & MAP_FIXED)?0:1, prot, prot, 0);
	if (rv != KERN_SUCCESS) {
		vm_object_deallocate(shm_handle->shm_object);
		error = ENOMEM;
		goto done2;
	}
	vm_map_inherit(&p->p_vmspace->vm_map,
		attach_va, attach_va + size, VM_INHERIT_SHARE);

	shmmap_s->va = attach_va;
	shmmap_s->shmid = shmid;
	shmseg->shm_lpid = p->p_pid;
	shmseg->shm_atime = time_second;
	shmseg->shm_nattch++;
	td->td_retval[0] = attach_va;
done2:
	mtx_unlock(&Giant);
	return (error);
}

int
shmat(td, uap)
	struct thread *td;
	struct shmat_args *uap;
{
	return kern_shmat(td, uap->shmid, uap->shmaddr, uap->shmflg);
}

struct oshmid_ds {
	struct	ipc_perm shm_perm;	/* operation perms */
	int	shm_segsz;		/* size of segment (bytes) */
	u_short	shm_cpid;		/* pid, creator */
	u_short	shm_lpid;		/* pid, last operation */
	short	shm_nattch;		/* no. of current attaches */
	time_t	shm_atime;		/* last attach time */
	time_t	shm_dtime;		/* last detach time */
	time_t	shm_ctime;		/* last change time */
	void	*shm_handle;		/* internal handle for shm segment */
};

struct oshmctl_args {
	int shmid;
	int cmd;
	struct oshmid_ds *ubuf;
};

/*
 * MPSAFE
 */
static int
oshmctl(td, uap)
	struct thread *td;
	struct oshmctl_args *uap;
{
#ifdef COMPAT_43
	int error = 0;
	struct shmid_ds *shmseg;
	struct oshmid_ds outbuf;

	if (!jail_sysvipc_allowed && jailed(td->td_ucred))
		return (ENOSYS);
	mtx_lock(&Giant);
	shmseg = shm_find_segment_by_shmid(uap->shmid);
	if (shmseg == NULL) {
		error = EINVAL;
		goto done2;
	}
	switch (uap->cmd) {
	case IPC_STAT:
		error = ipcperm(td, &shmseg->shm_perm, IPC_R);
		if (error)
			goto done2;
		outbuf.shm_perm = shmseg->shm_perm;
		outbuf.shm_segsz = shmseg->shm_segsz;
		outbuf.shm_cpid = shmseg->shm_cpid;
		outbuf.shm_lpid = shmseg->shm_lpid;
		outbuf.shm_nattch = shmseg->shm_nattch;
		outbuf.shm_atime = shmseg->shm_atime;
		outbuf.shm_dtime = shmseg->shm_dtime;
		outbuf.shm_ctime = shmseg->shm_ctime;
		outbuf.shm_handle = shmseg->shm_internal;
		error = copyout(&outbuf, uap->ubuf, sizeof(outbuf));
		if (error)
			goto done2;
		break;
	default:
		/* XXX casting to (sy_call_t *) is bogus, as usual. */
		error = ((sy_call_t *)shmctl)(td, uap);
		break;
	}
done2:
	mtx_unlock(&Giant);
	return (error);
#else
	return (EINVAL);
#endif
}

#ifndef _SYS_SYSPROTO_H_
struct shmctl_args {
	int shmid;
	int cmd;
	struct shmid_ds *buf;
};
#endif

/*
 * MPSAFE
 */
int
kern_shmctl(td, shmid, cmd, buf, bufsz)
	struct thread *td;
	int shmid;
	int cmd;
	void *buf;
	size_t *bufsz;
{
	int error = 0;
	struct shmid_ds *shmseg;

	if (!jail_sysvipc_allowed && jailed(td->td_ucred))
		return (ENOSYS);

	mtx_lock(&Giant);
	switch (cmd) {
	case IPC_INFO:
		memcpy(buf, &shminfo, sizeof(shminfo));
		if (bufsz)
			*bufsz = sizeof(shminfo);
		td->td_retval[0] = shmalloced;
		goto done2;
	case SHM_INFO: {
		struct shm_info shm_info;
		shm_info.used_ids = shm_nused;
		shm_info.shm_rss = 0;	/*XXX where to get from ? */
		shm_info.shm_tot = 0;	/*XXX where to get from ? */
		shm_info.shm_swp = 0;	/*XXX where to get from ? */
		shm_info.swap_attempts = 0;	/*XXX where to get from ? */
		shm_info.swap_successes = 0;	/*XXX where to get from ? */
		memcpy(buf, &shm_info, sizeof(shm_info));
		if (bufsz)
			*bufsz = sizeof(shm_info);
		td->td_retval[0] = shmalloced;
		goto done2;
	}
	}
	if (cmd == SHM_STAT)
		shmseg = shm_find_segment_by_shmidx(shmid);
	else
		shmseg = shm_find_segment_by_shmid(shmid);
	if (shmseg == NULL) {
		error = EINVAL;
		goto done2;
	}
	switch (cmd) {
	case SHM_STAT:
	case IPC_STAT:
		error = ipcperm(td, &shmseg->shm_perm, IPC_R);
		if (error)
			goto done2;
		memcpy(buf, shmseg, sizeof(struct shmid_ds));
		if (bufsz)
			*bufsz = sizeof(struct shmid_ds);
		if (cmd == SHM_STAT)
			td->td_retval[0] = IXSEQ_TO_IPCID(shmid, shmseg->shm_perm);
		break;
	case IPC_SET: {
		struct shmid_ds *shmid;

		shmid = (struct shmid_ds *)buf;
		error = ipcperm(td, &shmseg->shm_perm, IPC_M);
		if (error)
			goto done2;
		shmseg->shm_perm.uid = shmid->shm_perm.uid;
		shmseg->shm_perm.gid = shmid->shm_perm.gid;
		shmseg->shm_perm.mode =
		    (shmseg->shm_perm.mode & ~ACCESSPERMS) |
		    (shmid->shm_perm.mode & ACCESSPERMS);
		shmseg->shm_ctime = time_second;
		break;
	}
	case IPC_RMID:
		error = ipcperm(td, &shmseg->shm_perm, IPC_M);
		if (error)
			goto done2;
		shmseg->shm_perm.key = IPC_PRIVATE;
		shmseg->shm_perm.mode |= SHMSEG_REMOVED;
		if (shmseg->shm_nattch <= 0) {
			shm_deallocate_segment(shmseg);
			shm_last_free = IPCID_TO_IX(shmid);
		}
		break;
#if 0
	case SHM_LOCK:
	case SHM_UNLOCK:
#endif
	default:
		error = EINVAL;
		break;
	}
done2:
	mtx_unlock(&Giant);
	return (error);
}

int
shmctl(td, uap)
	struct thread *td;
	struct shmctl_args *uap;
{
	int error = 0;
	struct shmid_ds buf;
	size_t bufsz;
	
	/* IPC_SET needs to copyin the buffer before calling kern_shmctl */
	if (uap->cmd == IPC_SET) {
		if ((error = copyin(uap->buf, &buf, sizeof(struct shmid_ds))))
			goto done;
	}
	
	error = kern_shmctl(td, uap->shmid, uap->cmd, (void *)&buf, &bufsz);
	if (error)
		goto done;
	
	/* Cases in which we need to copyout */
	switch (uap->cmd) {
	case IPC_INFO:
	case SHM_INFO:
	case SHM_STAT:
	case IPC_STAT:
		error = copyout(&buf, uap->buf, bufsz);
		break;
	}

done:
	if (error) {
		/* Invalidate the return value */
		td->td_retval[0] = -1;
	}
	return (error);
}


#ifndef _SYS_SYSPROTO_H_
struct shmget_args {
	key_t key;
	size_t size;
	int shmflg;
};
#endif

static int
shmget_existing(td, uap, mode, segnum)
	struct thread *td;
	struct shmget_args *uap;
	int mode;
	int segnum;
{
	struct shmid_ds *shmseg;
	int error;

	shmseg = &shmsegs[segnum];
	if (shmseg->shm_perm.mode & SHMSEG_REMOVED) {
		/*
		 * This segment is in the process of being allocated.  Wait
		 * until it's done, and look the key up again (in case the
		 * allocation failed or it was freed).
		 */
		shmseg->shm_perm.mode |= SHMSEG_WANTED;
		error = tsleep(shmseg, PLOCK | PCATCH, "shmget", 0);
		if (error)
			return (error);
		return (EAGAIN);
	}
	if ((uap->shmflg & (IPC_CREAT | IPC_EXCL)) == (IPC_CREAT | IPC_EXCL))
		return (EEXIST);
	error = ipcperm(td, &shmseg->shm_perm, mode);
	if (error)
		return (error);
	if (uap->size && uap->size > shmseg->shm_segsz)
		return (EINVAL);
	td->td_retval[0] = IXSEQ_TO_IPCID(segnum, shmseg->shm_perm);
	return (0);
}

static int
shmget_allocate_segment(td, uap, mode)
	struct thread *td;
	struct shmget_args *uap;
	int mode;
{
	int i, segnum, shmid, size;
	struct ucred *cred = td->td_ucred;
	struct shmid_ds *shmseg;
	struct shm_handle *shm_handle;

	GIANT_REQUIRED;

	if (uap->size < shminfo.shmmin || uap->size > shminfo.shmmax)
		return (EINVAL);
	if (shm_nused >= shminfo.shmmni) /* Any shmids left? */
		return (ENOSPC);
	size = round_page(uap->size);
	if (shm_committed + btoc(size) > shminfo.shmall)
		return (ENOMEM);
	if (shm_last_free < 0) {
		shmrealloc();	/* Maybe expand the shmsegs[] array. */
		for (i = 0; i < shmalloced; i++)
			if (shmsegs[i].shm_perm.mode & SHMSEG_FREE)
				break;
		if (i == shmalloced)
			return (ENOSPC);
		segnum = i;
	} else  {
		segnum = shm_last_free;
		shm_last_free = -1;
	}
	shmseg = &shmsegs[segnum];
	/*
	 * In case we sleep in malloc(), mark the segment present but deleted
	 * so that noone else tries to create the same key.
	 */
	shmseg->shm_perm.mode = SHMSEG_ALLOCATED | SHMSEG_REMOVED;
	shmseg->shm_perm.key = uap->key;
	shmseg->shm_perm.seq = (shmseg->shm_perm.seq + 1) & 0x7fff;
	shm_handle = (struct shm_handle *)
	    malloc(sizeof(struct shm_handle), M_SHM, M_WAITOK);
	shmid = IXSEQ_TO_IPCID(segnum, shmseg->shm_perm);
	
	/*
	 * We make sure that we have allocated a pager before we need
	 * to.
	 */
	if (shm_use_phys) {
		shm_handle->shm_object =
		    vm_pager_allocate(OBJT_PHYS, 0, size, VM_PROT_DEFAULT, 0);
	} else {
		shm_handle->shm_object =
		    vm_pager_allocate(OBJT_SWAP, 0, size, VM_PROT_DEFAULT, 0);
	}
	VM_OBJECT_LOCK(shm_handle->shm_object);
	vm_object_clear_flag(shm_handle->shm_object, OBJ_ONEMAPPING);
	vm_object_set_flag(shm_handle->shm_object, OBJ_NOSPLIT);
	VM_OBJECT_UNLOCK(shm_handle->shm_object);

	shmseg->shm_internal = shm_handle;
	shmseg->shm_perm.cuid = shmseg->shm_perm.uid = cred->cr_uid;
	shmseg->shm_perm.cgid = shmseg->shm_perm.gid = cred->cr_gid;
	shmseg->shm_perm.mode = (shmseg->shm_perm.mode & SHMSEG_WANTED) |
	    (mode & ACCESSPERMS) | SHMSEG_ALLOCATED;
	shmseg->shm_segsz = uap->size;
	shmseg->shm_cpid = td->td_proc->p_pid;
	shmseg->shm_lpid = shmseg->shm_nattch = 0;
	shmseg->shm_atime = shmseg->shm_dtime = 0;
	shmseg->shm_ctime = time_second;
	shm_committed += btoc(size);
	shm_nused++;
	if (shmseg->shm_perm.mode & SHMSEG_WANTED) {
		/*
		 * Somebody else wanted this key while we were asleep.  Wake
		 * them up now.
		 */
		shmseg->shm_perm.mode &= ~SHMSEG_WANTED;
		wakeup(shmseg);
	}
	td->td_retval[0] = shmid;
	return (0);
}

/*
 * MPSAFE
 */
int
shmget(td, uap)
	struct thread *td;
	struct shmget_args *uap;
{
	int segnum, mode;
	int error;

	if (!jail_sysvipc_allowed && jailed(td->td_ucred))
		return (ENOSYS);
	mtx_lock(&Giant);
	mode = uap->shmflg & ACCESSPERMS;
	if (uap->key != IPC_PRIVATE) {
	again:
		segnum = shm_find_segment_by_key(uap->key);
		if (segnum >= 0) {
			error = shmget_existing(td, uap, mode, segnum);
			if (error == EAGAIN)
				goto again;
			goto done2;
		}
		if ((uap->shmflg & IPC_CREAT) == 0) {
			error = ENOENT;
			goto done2;
		}
	}
	error = shmget_allocate_segment(td, uap, mode);
done2:
	mtx_unlock(&Giant);
	return (error);
}

/*
 * MPSAFE
 */
int
shmsys(td, uap)
	struct thread *td;
	/* XXX actually varargs. */
	struct shmsys_args /* {
		int	which;
		int	a2;
		int	a3;
		int	a4;
	} */ *uap;
{
	int error;

	if (!jail_sysvipc_allowed && jailed(td->td_ucred))
		return (ENOSYS);
	if (uap->which < 0 ||
	    uap->which >= sizeof(shmcalls)/sizeof(shmcalls[0]))
		return (EINVAL);
	mtx_lock(&Giant);
	error = (*shmcalls[uap->which])(td, &uap->a2);
	mtx_unlock(&Giant);
	return (error);
}

static void
shmfork_myhook(p1, p2)
	struct proc *p1, *p2;
{
	struct shmmap_state *shmmap_s;
	size_t size;
	int i;

	size = shminfo.shmseg * sizeof(struct shmmap_state);
	shmmap_s = malloc(size, M_SHM, M_WAITOK);
	bcopy(p1->p_vmspace->vm_shm, shmmap_s, size);
	p2->p_vmspace->vm_shm = shmmap_s;
	for (i = 0; i < shminfo.shmseg; i++, shmmap_s++)
		if (shmmap_s->shmid != -1)
			shmsegs[IPCID_TO_IX(shmmap_s->shmid)].shm_nattch++;
}

static void
shmexit_myhook(struct vmspace *vm)
{
	struct shmmap_state *base, *shm;
	int i;

	GIANT_REQUIRED;

	if ((base = vm->vm_shm) != NULL) {
		vm->vm_shm = NULL;
		for (i = 0, shm = base; i < shminfo.shmseg; i++, shm++) {
			if (shm->shmid != -1)
				shm_delete_mapping(vm, shm);
		}
		free(base, M_SHM);
	}
}

static void
shmrealloc(void)
{
	int i;
	struct shmid_ds *newsegs;

	if (shmalloced >= shminfo.shmmni)
		return;

	newsegs = malloc(shminfo.shmmni * sizeof(*newsegs), M_SHM, M_WAITOK);
	if (newsegs == NULL)
		return;
	for (i = 0; i < shmalloced; i++)
		bcopy(&shmsegs[i], &newsegs[i], sizeof(newsegs[0]));
	for (; i < shminfo.shmmni; i++) {
		shmsegs[i].shm_perm.mode = SHMSEG_FREE;
		shmsegs[i].shm_perm.seq = 0;
	}
	free(shmsegs, M_SHM);
	shmsegs = newsegs;
	shmalloced = shminfo.shmmni;
}

static void
shminit()
{
	int i;

	TUNABLE_INT_FETCH("kern.ipc.shmmaxpgs", &shminfo.shmall);
	for (i = PAGE_SIZE; i > 0; i--) {
		shminfo.shmmax = shminfo.shmall * PAGE_SIZE;
		if (shminfo.shmmax >= shminfo.shmall)
			break;
	}
	TUNABLE_INT_FETCH("kern.ipc.shmmin", &shminfo.shmmin);
	TUNABLE_INT_FETCH("kern.ipc.shmmni", &shminfo.shmmni);
	TUNABLE_INT_FETCH("kern.ipc.shmseg", &shminfo.shmseg);
	TUNABLE_INT_FETCH("kern.ipc.shm_use_phys", &shm_use_phys);

	shmalloced = shminfo.shmmni;
	shmsegs = malloc(shmalloced * sizeof(shmsegs[0]), M_SHM, M_WAITOK);
	if (shmsegs == NULL)
		panic("cannot allocate initial memory for sysvshm");
	for (i = 0; i < shmalloced; i++) {
		shmsegs[i].shm_perm.mode = SHMSEG_FREE;
		shmsegs[i].shm_perm.seq = 0;
	}
	shm_last_free = 0;
	shm_nused = 0;
	shm_committed = 0;
	shmexit_hook = &shmexit_myhook;
	shmfork_hook = &shmfork_myhook;
}

static int
shmunload()
{

	if (shm_nused > 0)
		return (EBUSY);

	free(shmsegs, M_SHM);
	shmexit_hook = NULL;
	shmfork_hook = NULL;
	return (0);
}

static int
sysctl_shmsegs(SYSCTL_HANDLER_ARGS)
{

	return (SYSCTL_OUT(req, shmsegs, shmalloced * sizeof(shmsegs[0])));
}

static int
sysvshm_modload(struct module *module, int cmd, void *arg)
{
	int error = 0;

	switch (cmd) {
	case MOD_LOAD:
		shminit();
		break;
	case MOD_UNLOAD:
		error = shmunload();
		break;
	case MOD_SHUTDOWN:
		break;
	default:
		error = EINVAL;
		break;
	}
	return (error);
}

static moduledata_t sysvshm_mod = {
	"sysvshm",
	&sysvshm_modload,
	NULL
};

SYSCALL_MODULE_HELPER(shmsys);
SYSCALL_MODULE_HELPER(shmat);
SYSCALL_MODULE_HELPER(shmctl);
SYSCALL_MODULE_HELPER(shmdt);
SYSCALL_MODULE_HELPER(shmget);

DECLARE_MODULE(sysvshm, sysvshm_mod,
	SI_SUB_SYSV_SHM, SI_ORDER_FIRST);
MODULE_VERSION(sysvshm, 1);
OpenPOWER on IntegriCloud