xref: /freebsd/sys/kern/sysv_shm.c (revision cc16dea626cf2fc80cde667ac4798065108e596c)
1 /*	$NetBSD: sysv_shm.c,v 1.23 1994/07/04 23:25:12 glass Exp $	*/
2 /*-
3  * Copyright (c) 1994 Adam Glass and Charles Hannum.  All rights reserved.
4  *
5  * Redistribution and use in source and binary forms, with or without
6  * modification, are permitted provided that the following conditions
7  * are met:
8  * 1. Redistributions of source code must retain the above copyright
9  *    notice, this list of conditions and the following disclaimer.
10  * 2. Redistributions in binary form must reproduce the above copyright
11  *    notice, this list of conditions and the following disclaimer in the
12  *    documentation and/or other materials provided with the distribution.
13  * 3. All advertising materials mentioning features or use of this software
14  *    must display the following acknowledgement:
15  *	This product includes software developed by Adam Glass and Charles
16  *	Hannum.
17  * 4. The names of the authors may not be used to endorse or promote products
18  *    derived from this software without specific prior written permission.
19  *
20  * THIS SOFTWARE IS PROVIDED BY THE AUTHORS ``AS IS'' AND ANY EXPRESS OR
21  * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
22  * OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
23  * IN NO EVENT SHALL THE AUTHORS BE LIABLE FOR ANY DIRECT, INDIRECT,
24  * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
25  * NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
26  * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
27  * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
28  * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
29  * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
30  */
31 /*-
32  * Copyright (c) 2003-2005 McAfee, Inc.
33  * All rights reserved.
34  *
35  * This software was developed for the FreeBSD Project in part by McAfee
36  * Research, the Security Research Division of McAfee, Inc under DARPA/SPAWAR
37  * contract N66001-01-C-8035 ("CBOSS"), as part of the DARPA CHATS research
38  * program.
39  *
40  * Redistribution and use in source and binary forms, with or without
41  * modification, are permitted provided that the following conditions
42  * are met:
43  * 1. Redistributions of source code must retain the above copyright
44  *    notice, this list of conditions and the following disclaimer.
45  * 2. Redistributions in binary form must reproduce the above copyright
46  *    notice, this list of conditions and the following disclaimer in the
47  *    documentation and/or other materials provided with the distribution.
48  *
49  * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
50  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
51  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
52  * ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
53  * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
54  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
55  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
56  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
57  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
58  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
59  * SUCH DAMAGE.
60  */
61 
62 #include <sys/cdefs.h>
63 __FBSDID("$FreeBSD$");
64 
65 #include "opt_compat.h"
66 #include "opt_sysvipc.h"
67 
68 #include <sys/param.h>
69 #include <sys/systm.h>
70 #include <sys/kernel.h>
71 #include <sys/limits.h>
72 #include <sys/lock.h>
73 #include <sys/sysctl.h>
74 #include <sys/shm.h>
75 #include <sys/proc.h>
76 #include <sys/malloc.h>
77 #include <sys/mman.h>
78 #include <sys/module.h>
79 #include <sys/mutex.h>
80 #include <sys/racct.h>
81 #include <sys/resourcevar.h>
82 #include <sys/rwlock.h>
83 #include <sys/stat.h>
84 #include <sys/syscall.h>
85 #include <sys/syscallsubr.h>
86 #include <sys/sysent.h>
87 #include <sys/sysproto.h>
88 #include <sys/jail.h>
89 
90 #include <security/mac/mac_framework.h>
91 
92 #include <vm/vm.h>
93 #include <vm/vm_param.h>
94 #include <vm/pmap.h>
95 #include <vm/vm_object.h>
96 #include <vm/vm_map.h>
97 #include <vm/vm_page.h>
98 #include <vm/vm_pager.h>
99 
100 FEATURE(sysv_shm, "System V shared memory segments support");
101 
102 static MALLOC_DEFINE(M_SHM, "shm", "SVID compatible shared memory segments");
103 
104 static int shmget_allocate_segment(struct thread *td,
105     struct shmget_args *uap, int mode);
106 static int shmget_existing(struct thread *td, struct shmget_args *uap,
107     int mode, int segnum);
108 
109 #define	SHMSEG_FREE     	0x0200
110 #define	SHMSEG_REMOVED  	0x0400
111 #define	SHMSEG_ALLOCATED	0x0800
112 #define	SHMSEG_WANTED		0x1000
113 
114 static int shm_last_free, shm_nused, shmalloced;
115 vm_size_t shm_committed;
116 static struct shmid_kernel	*shmsegs;
117 
118 struct shmmap_state {
119 	vm_offset_t va;
120 	int shmid;
121 };
122 
123 static void shm_deallocate_segment(struct shmid_kernel *);
124 static int shm_find_segment_by_key(key_t);
125 static struct shmid_kernel *shm_find_segment_by_shmid(int);
126 static struct shmid_kernel *shm_find_segment_by_shmidx(int);
127 static int shm_delete_mapping(struct vmspace *vm, struct shmmap_state *);
128 static void shmrealloc(void);
129 static int shminit(void);
130 static int sysvshm_modload(struct module *, int, void *);
131 static int shmunload(void);
132 static void shmexit_myhook(struct vmspace *vm);
133 static void shmfork_myhook(struct proc *p1, struct proc *p2);
134 static int sysctl_shmsegs(SYSCTL_HANDLER_ARGS);
135 
136 /*
137  * Tuneable values.
138  */
139 #ifndef SHMMAXPGS
140 #define	SHMMAXPGS	131072	/* Note: sysv shared memory is swap backed. */
141 #endif
142 #ifndef SHMMAX
143 #define	SHMMAX	(SHMMAXPGS*PAGE_SIZE)
144 #endif
145 #ifndef SHMMIN
146 #define	SHMMIN	1
147 #endif
148 #ifndef SHMMNI
149 #define	SHMMNI	192
150 #endif
151 #ifndef SHMSEG
152 #define	SHMSEG	128
153 #endif
154 #ifndef SHMALL
155 #define	SHMALL	(SHMMAXPGS)
156 #endif
157 
158 struct	shminfo shminfo = {
159 	SHMMAX,
160 	SHMMIN,
161 	SHMMNI,
162 	SHMSEG,
163 	SHMALL
164 };
165 
166 static int shm_use_phys;
167 static int shm_allow_removed;
168 
169 SYSCTL_ULONG(_kern_ipc, OID_AUTO, shmmax, CTLFLAG_RW, &shminfo.shmmax, 0,
170     "Maximum shared memory segment size");
171 SYSCTL_ULONG(_kern_ipc, OID_AUTO, shmmin, CTLFLAG_RW, &shminfo.shmmin, 0,
172     "Minimum shared memory segment size");
173 SYSCTL_ULONG(_kern_ipc, OID_AUTO, shmmni, CTLFLAG_RDTUN, &shminfo.shmmni, 0,
174     "Number of shared memory identifiers");
175 SYSCTL_ULONG(_kern_ipc, OID_AUTO, shmseg, CTLFLAG_RDTUN, &shminfo.shmseg, 0,
176     "Number of segments per process");
177 SYSCTL_ULONG(_kern_ipc, OID_AUTO, shmall, CTLFLAG_RW, &shminfo.shmall, 0,
178     "Maximum number of pages available for shared memory");
179 SYSCTL_INT(_kern_ipc, OID_AUTO, shm_use_phys, CTLFLAG_RW,
180     &shm_use_phys, 0, "Enable/Disable locking of shared memory pages in core");
181 SYSCTL_INT(_kern_ipc, OID_AUTO, shm_allow_removed, CTLFLAG_RW,
182     &shm_allow_removed, 0,
183     "Enable/Disable attachment to attached segments marked for removal");
184 SYSCTL_PROC(_kern_ipc, OID_AUTO, shmsegs, CTLTYPE_OPAQUE | CTLFLAG_RD,
185     NULL, 0, sysctl_shmsegs, "",
186     "Current number of shared memory segments allocated");
187 
188 static int
189 shm_find_segment_by_key(key)
190 	key_t key;
191 {
192 	int i;
193 
194 	for (i = 0; i < shmalloced; i++)
195 		if ((shmsegs[i].u.shm_perm.mode & SHMSEG_ALLOCATED) &&
196 		    shmsegs[i].u.shm_perm.key == key)
197 			return (i);
198 	return (-1);
199 }
200 
201 static struct shmid_kernel *
202 shm_find_segment_by_shmid(int shmid)
203 {
204 	int segnum;
205 	struct shmid_kernel *shmseg;
206 
207 	segnum = IPCID_TO_IX(shmid);
208 	if (segnum < 0 || segnum >= shmalloced)
209 		return (NULL);
210 	shmseg = &shmsegs[segnum];
211 	if ((shmseg->u.shm_perm.mode & SHMSEG_ALLOCATED) == 0 ||
212 	    (!shm_allow_removed &&
213 	     (shmseg->u.shm_perm.mode & SHMSEG_REMOVED) != 0) ||
214 	    shmseg->u.shm_perm.seq != IPCID_TO_SEQ(shmid))
215 		return (NULL);
216 	return (shmseg);
217 }
218 
219 static struct shmid_kernel *
220 shm_find_segment_by_shmidx(int segnum)
221 {
222 	struct shmid_kernel *shmseg;
223 
224 	if (segnum < 0 || segnum >= shmalloced)
225 		return (NULL);
226 	shmseg = &shmsegs[segnum];
227 	if ((shmseg->u.shm_perm.mode & SHMSEG_ALLOCATED) == 0 ||
228 	    (!shm_allow_removed &&
229 	     (shmseg->u.shm_perm.mode & SHMSEG_REMOVED) != 0))
230 		return (NULL);
231 	return (shmseg);
232 }
233 
234 static void
235 shm_deallocate_segment(shmseg)
236 	struct shmid_kernel *shmseg;
237 {
238 	vm_size_t size;
239 
240 	GIANT_REQUIRED;
241 
242 	vm_object_deallocate(shmseg->object);
243 	shmseg->object = NULL;
244 	size = round_page(shmseg->u.shm_segsz);
245 	shm_committed -= btoc(size);
246 	shm_nused--;
247 	shmseg->u.shm_perm.mode = SHMSEG_FREE;
248 #ifdef MAC
249 	mac_sysvshm_cleanup(shmseg);
250 #endif
251 	racct_sub_cred(shmseg->cred, RACCT_NSHM, 1);
252 	racct_sub_cred(shmseg->cred, RACCT_SHMSIZE, size);
253 	crfree(shmseg->cred);
254 	shmseg->cred = NULL;
255 }
256 
257 static int
258 shm_delete_mapping(struct vmspace *vm, struct shmmap_state *shmmap_s)
259 {
260 	struct shmid_kernel *shmseg;
261 	int segnum, result;
262 	vm_size_t size;
263 
264 	GIANT_REQUIRED;
265 
266 	segnum = IPCID_TO_IX(shmmap_s->shmid);
267 	shmseg = &shmsegs[segnum];
268 	size = round_page(shmseg->u.shm_segsz);
269 	result = vm_map_remove(&vm->vm_map, shmmap_s->va, shmmap_s->va + size);
270 	if (result != KERN_SUCCESS)
271 		return (EINVAL);
272 	shmmap_s->shmid = -1;
273 	shmseg->u.shm_dtime = time_second;
274 	if ((--shmseg->u.shm_nattch <= 0) &&
275 	    (shmseg->u.shm_perm.mode & SHMSEG_REMOVED)) {
276 		shm_deallocate_segment(shmseg);
277 		shm_last_free = segnum;
278 	}
279 	return (0);
280 }
281 
282 #ifndef _SYS_SYSPROTO_H_
283 struct shmdt_args {
284 	const void *shmaddr;
285 };
286 #endif
287 int
288 sys_shmdt(td, uap)
289 	struct thread *td;
290 	struct shmdt_args *uap;
291 {
292 	struct proc *p = td->td_proc;
293 	struct shmmap_state *shmmap_s;
294 #ifdef MAC
295 	struct shmid_kernel *shmsegptr;
296 #endif
297 	int i;
298 	int error = 0;
299 
300 	if (!prison_allow(td->td_ucred, PR_ALLOW_SYSVIPC))
301 		return (ENOSYS);
302 	mtx_lock(&Giant);
303 	shmmap_s = p->p_vmspace->vm_shm;
304  	if (shmmap_s == NULL) {
305 		error = EINVAL;
306 		goto done2;
307 	}
308 	for (i = 0; i < shminfo.shmseg; i++, shmmap_s++) {
309 		if (shmmap_s->shmid != -1 &&
310 		    shmmap_s->va == (vm_offset_t)uap->shmaddr) {
311 			break;
312 		}
313 	}
314 	if (i == shminfo.shmseg) {
315 		error = EINVAL;
316 		goto done2;
317 	}
318 #ifdef MAC
319 	shmsegptr = &shmsegs[IPCID_TO_IX(shmmap_s->shmid)];
320 	error = mac_sysvshm_check_shmdt(td->td_ucred, shmsegptr);
321 	if (error != 0)
322 		goto done2;
323 #endif
324 	error = shm_delete_mapping(p->p_vmspace, shmmap_s);
325 done2:
326 	mtx_unlock(&Giant);
327 	return (error);
328 }
329 
330 #ifndef _SYS_SYSPROTO_H_
331 struct shmat_args {
332 	int shmid;
333 	const void *shmaddr;
334 	int shmflg;
335 };
336 #endif
337 int
338 kern_shmat(td, shmid, shmaddr, shmflg)
339 	struct thread *td;
340 	int shmid;
341 	const void *shmaddr;
342 	int shmflg;
343 {
344 	struct proc *p = td->td_proc;
345 	int i, flags;
346 	struct shmid_kernel *shmseg;
347 	struct shmmap_state *shmmap_s = NULL;
348 	vm_offset_t attach_va;
349 	vm_prot_t prot;
350 	vm_size_t size;
351 	int rv;
352 	int error = 0;
353 
354 	if (!prison_allow(td->td_ucred, PR_ALLOW_SYSVIPC))
355 		return (ENOSYS);
356 	mtx_lock(&Giant);
357 	shmmap_s = p->p_vmspace->vm_shm;
358 	if (shmmap_s == NULL) {
359 		shmmap_s = malloc(shminfo.shmseg * sizeof(struct shmmap_state),
360 		    M_SHM, M_WAITOK);
361 		for (i = 0; i < shminfo.shmseg; i++)
362 			shmmap_s[i].shmid = -1;
363 		p->p_vmspace->vm_shm = shmmap_s;
364 	}
365 	shmseg = shm_find_segment_by_shmid(shmid);
366 	if (shmseg == NULL) {
367 		error = EINVAL;
368 		goto done2;
369 	}
370 	error = ipcperm(td, &shmseg->u.shm_perm,
371 	    (shmflg & SHM_RDONLY) ? IPC_R : IPC_R|IPC_W);
372 	if (error)
373 		goto done2;
374 #ifdef MAC
375 	error = mac_sysvshm_check_shmat(td->td_ucred, shmseg, shmflg);
376 	if (error != 0)
377 		goto done2;
378 #endif
379 	for (i = 0; i < shminfo.shmseg; i++) {
380 		if (shmmap_s->shmid == -1)
381 			break;
382 		shmmap_s++;
383 	}
384 	if (i >= shminfo.shmseg) {
385 		error = EMFILE;
386 		goto done2;
387 	}
388 	size = round_page(shmseg->u.shm_segsz);
389 	prot = VM_PROT_READ;
390 	if ((shmflg & SHM_RDONLY) == 0)
391 		prot |= VM_PROT_WRITE;
392 	flags = MAP_ANON | MAP_SHARED;
393 	if (shmaddr) {
394 		flags |= MAP_FIXED;
395 		if (shmflg & SHM_RND) {
396 			attach_va = (vm_offset_t)shmaddr & ~(SHMLBA-1);
397 		} else if (((vm_offset_t)shmaddr & (SHMLBA-1)) == 0) {
398 			attach_va = (vm_offset_t)shmaddr;
399 		} else {
400 			error = EINVAL;
401 			goto done2;
402 		}
403 	} else {
404 		/*
405 		 * This is just a hint to vm_map_find() about where to
406 		 * put it.
407 		 */
408 		PROC_LOCK(p);
409 		attach_va = round_page((vm_offset_t)p->p_vmspace->vm_daddr +
410 		    lim_max(p, RLIMIT_DATA));
411 		PROC_UNLOCK(p);
412 	}
413 
414 	vm_object_reference(shmseg->object);
415 	rv = vm_map_find(&p->p_vmspace->vm_map, shmseg->object,
416 	    0, &attach_va, size, (flags & MAP_FIXED) ? VMFS_NO_SPACE :
417 	    VMFS_OPTIMAL_SPACE, prot, prot, MAP_INHERIT_SHARE);
418 	if (rv != KERN_SUCCESS) {
419 		vm_object_deallocate(shmseg->object);
420 		error = ENOMEM;
421 		goto done2;
422 	}
423 
424 	shmmap_s->va = attach_va;
425 	shmmap_s->shmid = shmid;
426 	shmseg->u.shm_lpid = p->p_pid;
427 	shmseg->u.shm_atime = time_second;
428 	shmseg->u.shm_nattch++;
429 	td->td_retval[0] = attach_va;
430 done2:
431 	mtx_unlock(&Giant);
432 	return (error);
433 }
434 
435 int
436 sys_shmat(td, uap)
437 	struct thread *td;
438 	struct shmat_args *uap;
439 {
440 	return kern_shmat(td, uap->shmid, uap->shmaddr, uap->shmflg);
441 }
442 
443 int
444 kern_shmctl(td, shmid, cmd, buf, bufsz)
445 	struct thread *td;
446 	int shmid;
447 	int cmd;
448 	void *buf;
449 	size_t *bufsz;
450 {
451 	int error = 0;
452 	struct shmid_kernel *shmseg;
453 
454 	if (!prison_allow(td->td_ucred, PR_ALLOW_SYSVIPC))
455 		return (ENOSYS);
456 
457 	mtx_lock(&Giant);
458 	switch (cmd) {
459 	/*
460 	 * It is possible that kern_shmctl is being called from the Linux ABI
461 	 * layer, in which case, we will need to implement IPC_INFO.  It should
462 	 * be noted that other shmctl calls will be funneled through here for
463 	 * Linix binaries as well.
464 	 *
465 	 * NB: The Linux ABI layer will convert this data to structure(s) more
466 	 * consistent with the Linux ABI.
467 	 */
468 	case IPC_INFO:
469 		memcpy(buf, &shminfo, sizeof(shminfo));
470 		if (bufsz)
471 			*bufsz = sizeof(shminfo);
472 		td->td_retval[0] = shmalloced;
473 		goto done2;
474 	case SHM_INFO: {
475 		struct shm_info shm_info;
476 		shm_info.used_ids = shm_nused;
477 		shm_info.shm_rss = 0;	/*XXX where to get from ? */
478 		shm_info.shm_tot = 0;	/*XXX where to get from ? */
479 		shm_info.shm_swp = 0;	/*XXX where to get from ? */
480 		shm_info.swap_attempts = 0;	/*XXX where to get from ? */
481 		shm_info.swap_successes = 0;	/*XXX where to get from ? */
482 		memcpy(buf, &shm_info, sizeof(shm_info));
483 		if (bufsz)
484 			*bufsz = sizeof(shm_info);
485 		td->td_retval[0] = shmalloced;
486 		goto done2;
487 	}
488 	}
489 	if (cmd == SHM_STAT)
490 		shmseg = shm_find_segment_by_shmidx(shmid);
491 	else
492 		shmseg = shm_find_segment_by_shmid(shmid);
493 	if (shmseg == NULL) {
494 		error = EINVAL;
495 		goto done2;
496 	}
497 #ifdef MAC
498 	error = mac_sysvshm_check_shmctl(td->td_ucred, shmseg, cmd);
499 	if (error != 0)
500 		goto done2;
501 #endif
502 	switch (cmd) {
503 	case SHM_STAT:
504 	case IPC_STAT:
505 		error = ipcperm(td, &shmseg->u.shm_perm, IPC_R);
506 		if (error)
507 			goto done2;
508 		memcpy(buf, &shmseg->u, sizeof(struct shmid_ds));
509 		if (bufsz)
510 			*bufsz = sizeof(struct shmid_ds);
511 		if (cmd == SHM_STAT)
512 			td->td_retval[0] = IXSEQ_TO_IPCID(shmid, shmseg->u.shm_perm);
513 		break;
514 	case IPC_SET: {
515 		struct shmid_ds *shmid;
516 
517 		shmid = (struct shmid_ds *)buf;
518 		error = ipcperm(td, &shmseg->u.shm_perm, IPC_M);
519 		if (error)
520 			goto done2;
521 		shmseg->u.shm_perm.uid = shmid->shm_perm.uid;
522 		shmseg->u.shm_perm.gid = shmid->shm_perm.gid;
523 		shmseg->u.shm_perm.mode =
524 		    (shmseg->u.shm_perm.mode & ~ACCESSPERMS) |
525 		    (shmid->shm_perm.mode & ACCESSPERMS);
526 		shmseg->u.shm_ctime = time_second;
527 		break;
528 	}
529 	case IPC_RMID:
530 		error = ipcperm(td, &shmseg->u.shm_perm, IPC_M);
531 		if (error)
532 			goto done2;
533 		shmseg->u.shm_perm.key = IPC_PRIVATE;
534 		shmseg->u.shm_perm.mode |= SHMSEG_REMOVED;
535 		if (shmseg->u.shm_nattch <= 0) {
536 			shm_deallocate_segment(shmseg);
537 			shm_last_free = IPCID_TO_IX(shmid);
538 		}
539 		break;
540 #if 0
541 	case SHM_LOCK:
542 	case SHM_UNLOCK:
543 #endif
544 	default:
545 		error = EINVAL;
546 		break;
547 	}
548 done2:
549 	mtx_unlock(&Giant);
550 	return (error);
551 }
552 
553 #ifndef _SYS_SYSPROTO_H_
554 struct shmctl_args {
555 	int shmid;
556 	int cmd;
557 	struct shmid_ds *buf;
558 };
559 #endif
560 int
561 sys_shmctl(td, uap)
562 	struct thread *td;
563 	struct shmctl_args *uap;
564 {
565 	int error = 0;
566 	struct shmid_ds buf;
567 	size_t bufsz;
568 
569 	/*
570 	 * The only reason IPC_INFO, SHM_INFO, SHM_STAT exists is to support
571 	 * Linux binaries.  If we see the call come through the FreeBSD ABI,
572 	 * return an error back to the user since we do not to support this.
573 	 */
574 	if (uap->cmd == IPC_INFO || uap->cmd == SHM_INFO ||
575 	    uap->cmd == SHM_STAT)
576 		return (EINVAL);
577 
578 	/* IPC_SET needs to copyin the buffer before calling kern_shmctl */
579 	if (uap->cmd == IPC_SET) {
580 		if ((error = copyin(uap->buf, &buf, sizeof(struct shmid_ds))))
581 			goto done;
582 	}
583 
584 	error = kern_shmctl(td, uap->shmid, uap->cmd, (void *)&buf, &bufsz);
585 	if (error)
586 		goto done;
587 
588 	/* Cases in which we need to copyout */
589 	switch (uap->cmd) {
590 	case IPC_STAT:
591 		error = copyout(&buf, uap->buf, bufsz);
592 		break;
593 	}
594 
595 done:
596 	if (error) {
597 		/* Invalidate the return value */
598 		td->td_retval[0] = -1;
599 	}
600 	return (error);
601 }
602 
603 
604 static int
605 shmget_existing(td, uap, mode, segnum)
606 	struct thread *td;
607 	struct shmget_args *uap;
608 	int mode;
609 	int segnum;
610 {
611 	struct shmid_kernel *shmseg;
612 	int error;
613 
614 	shmseg = &shmsegs[segnum];
615 	if (shmseg->u.shm_perm.mode & SHMSEG_REMOVED) {
616 		/*
617 		 * This segment is in the process of being allocated.  Wait
618 		 * until it's done, and look the key up again (in case the
619 		 * allocation failed or it was freed).
620 		 */
621 		shmseg->u.shm_perm.mode |= SHMSEG_WANTED;
622 		error = tsleep(shmseg, PLOCK | PCATCH, "shmget", 0);
623 		if (error)
624 			return (error);
625 		return (EAGAIN);
626 	}
627 	if ((uap->shmflg & (IPC_CREAT | IPC_EXCL)) == (IPC_CREAT | IPC_EXCL))
628 		return (EEXIST);
629 #ifdef MAC
630 	error = mac_sysvshm_check_shmget(td->td_ucred, shmseg, uap->shmflg);
631 	if (error != 0)
632 		return (error);
633 #endif
634 	if (uap->size != 0 && uap->size > shmseg->u.shm_segsz)
635 		return (EINVAL);
636 	td->td_retval[0] = IXSEQ_TO_IPCID(segnum, shmseg->u.shm_perm);
637 	return (0);
638 }
639 
640 static int
641 shmget_allocate_segment(td, uap, mode)
642 	struct thread *td;
643 	struct shmget_args *uap;
644 	int mode;
645 {
646 	int i, segnum, shmid;
647 	size_t size;
648 	struct ucred *cred = td->td_ucred;
649 	struct shmid_kernel *shmseg;
650 	vm_object_t shm_object;
651 
652 	GIANT_REQUIRED;
653 
654 	if (uap->size < shminfo.shmmin || uap->size > shminfo.shmmax)
655 		return (EINVAL);
656 	if (shm_nused >= shminfo.shmmni) /* Any shmids left? */
657 		return (ENOSPC);
658 	size = round_page(uap->size);
659 	if (shm_committed + btoc(size) > shminfo.shmall)
660 		return (ENOMEM);
661 	if (shm_last_free < 0) {
662 		shmrealloc();	/* Maybe expand the shmsegs[] array. */
663 		for (i = 0; i < shmalloced; i++)
664 			if (shmsegs[i].u.shm_perm.mode & SHMSEG_FREE)
665 				break;
666 		if (i == shmalloced)
667 			return (ENOSPC);
668 		segnum = i;
669 	} else  {
670 		segnum = shm_last_free;
671 		shm_last_free = -1;
672 	}
673 	shmseg = &shmsegs[segnum];
674 #ifdef RACCT
675 	PROC_LOCK(td->td_proc);
676 	if (racct_add(td->td_proc, RACCT_NSHM, 1)) {
677 		PROC_UNLOCK(td->td_proc);
678 		return (ENOSPC);
679 	}
680 	if (racct_add(td->td_proc, RACCT_SHMSIZE, size)) {
681 		racct_sub(td->td_proc, RACCT_NSHM, 1);
682 		PROC_UNLOCK(td->td_proc);
683 		return (ENOMEM);
684 	}
685 	PROC_UNLOCK(td->td_proc);
686 #endif
687 	/*
688 	 * In case we sleep in malloc(), mark the segment present but deleted
689 	 * so that noone else tries to create the same key.
690 	 */
691 	shmseg->u.shm_perm.mode = SHMSEG_ALLOCATED | SHMSEG_REMOVED;
692 	shmseg->u.shm_perm.key = uap->key;
693 	shmseg->u.shm_perm.seq = (shmseg->u.shm_perm.seq + 1) & 0x7fff;
694 	shmid = IXSEQ_TO_IPCID(segnum, shmseg->u.shm_perm);
695 
696 	/*
697 	 * We make sure that we have allocated a pager before we need
698 	 * to.
699 	 */
700 	shm_object = vm_pager_allocate(shm_use_phys ? OBJT_PHYS : OBJT_SWAP,
701 	    0, size, VM_PROT_DEFAULT, 0, cred);
702 	if (shm_object == NULL) {
703 #ifdef RACCT
704 		PROC_LOCK(td->td_proc);
705 		racct_sub(td->td_proc, RACCT_NSHM, 1);
706 		racct_sub(td->td_proc, RACCT_SHMSIZE, size);
707 		PROC_UNLOCK(td->td_proc);
708 #endif
709 		return (ENOMEM);
710 	}
711 	VM_OBJECT_WLOCK(shm_object);
712 	vm_object_clear_flag(shm_object, OBJ_ONEMAPPING);
713 	vm_object_set_flag(shm_object, OBJ_NOSPLIT);
714 	VM_OBJECT_WUNLOCK(shm_object);
715 
716 	shmseg->object = shm_object;
717 	shmseg->u.shm_perm.cuid = shmseg->u.shm_perm.uid = cred->cr_uid;
718 	shmseg->u.shm_perm.cgid = shmseg->u.shm_perm.gid = cred->cr_gid;
719 	shmseg->u.shm_perm.mode = (shmseg->u.shm_perm.mode & SHMSEG_WANTED) |
720 	    (mode & ACCESSPERMS) | SHMSEG_ALLOCATED;
721 	shmseg->cred = crhold(cred);
722 	shmseg->u.shm_segsz = uap->size;
723 	shmseg->u.shm_cpid = td->td_proc->p_pid;
724 	shmseg->u.shm_lpid = shmseg->u.shm_nattch = 0;
725 	shmseg->u.shm_atime = shmseg->u.shm_dtime = 0;
726 #ifdef MAC
727 	mac_sysvshm_create(cred, shmseg);
728 #endif
729 	shmseg->u.shm_ctime = time_second;
730 	shm_committed += btoc(size);
731 	shm_nused++;
732 	if (shmseg->u.shm_perm.mode & SHMSEG_WANTED) {
733 		/*
734 		 * Somebody else wanted this key while we were asleep.  Wake
735 		 * them up now.
736 		 */
737 		shmseg->u.shm_perm.mode &= ~SHMSEG_WANTED;
738 		wakeup(shmseg);
739 	}
740 	td->td_retval[0] = shmid;
741 	return (0);
742 }
743 
744 #ifndef _SYS_SYSPROTO_H_
745 struct shmget_args {
746 	key_t key;
747 	size_t size;
748 	int shmflg;
749 };
750 #endif
751 int
752 sys_shmget(td, uap)
753 	struct thread *td;
754 	struct shmget_args *uap;
755 {
756 	int segnum, mode;
757 	int error;
758 
759 	if (!prison_allow(td->td_ucred, PR_ALLOW_SYSVIPC))
760 		return (ENOSYS);
761 	mtx_lock(&Giant);
762 	mode = uap->shmflg & ACCESSPERMS;
763 	if (uap->key != IPC_PRIVATE) {
764 	again:
765 		segnum = shm_find_segment_by_key(uap->key);
766 		if (segnum >= 0) {
767 			error = shmget_existing(td, uap, mode, segnum);
768 			if (error == EAGAIN)
769 				goto again;
770 			goto done2;
771 		}
772 		if ((uap->shmflg & IPC_CREAT) == 0) {
773 			error = ENOENT;
774 			goto done2;
775 		}
776 	}
777 	error = shmget_allocate_segment(td, uap, mode);
778 done2:
779 	mtx_unlock(&Giant);
780 	return (error);
781 }
782 
783 static void
784 shmfork_myhook(p1, p2)
785 	struct proc *p1, *p2;
786 {
787 	struct shmmap_state *shmmap_s;
788 	size_t size;
789 	int i;
790 
791 	mtx_lock(&Giant);
792 	size = shminfo.shmseg * sizeof(struct shmmap_state);
793 	shmmap_s = malloc(size, M_SHM, M_WAITOK);
794 	bcopy(p1->p_vmspace->vm_shm, shmmap_s, size);
795 	p2->p_vmspace->vm_shm = shmmap_s;
796 	for (i = 0; i < shminfo.shmseg; i++, shmmap_s++)
797 		if (shmmap_s->shmid != -1)
798 			shmsegs[IPCID_TO_IX(shmmap_s->shmid)].u.shm_nattch++;
799 	mtx_unlock(&Giant);
800 }
801 
802 static void
803 shmexit_myhook(struct vmspace *vm)
804 {
805 	struct shmmap_state *base, *shm;
806 	int i;
807 
808 	if ((base = vm->vm_shm) != NULL) {
809 		vm->vm_shm = NULL;
810 		mtx_lock(&Giant);
811 		for (i = 0, shm = base; i < shminfo.shmseg; i++, shm++) {
812 			if (shm->shmid != -1)
813 				shm_delete_mapping(vm, shm);
814 		}
815 		mtx_unlock(&Giant);
816 		free(base, M_SHM);
817 	}
818 }
819 
820 static void
821 shmrealloc(void)
822 {
823 	int i;
824 	struct shmid_kernel *newsegs;
825 
826 	if (shmalloced >= shminfo.shmmni)
827 		return;
828 
829 	newsegs = malloc(shminfo.shmmni * sizeof(*newsegs), M_SHM, M_WAITOK);
830 	if (newsegs == NULL)
831 		return;
832 	for (i = 0; i < shmalloced; i++)
833 		bcopy(&shmsegs[i], &newsegs[i], sizeof(newsegs[0]));
834 	for (; i < shminfo.shmmni; i++) {
835 		shmsegs[i].u.shm_perm.mode = SHMSEG_FREE;
836 		shmsegs[i].u.shm_perm.seq = 0;
837 #ifdef MAC
838 		mac_sysvshm_init(&shmsegs[i]);
839 #endif
840 	}
841 	free(shmsegs, M_SHM);
842 	shmsegs = newsegs;
843 	shmalloced = shminfo.shmmni;
844 }
845 
846 static struct syscall_helper_data shm_syscalls[] = {
847 	SYSCALL_INIT_HELPER(shmat),
848 	SYSCALL_INIT_HELPER(shmctl),
849 	SYSCALL_INIT_HELPER(shmdt),
850 	SYSCALL_INIT_HELPER(shmget),
851 #if defined(COMPAT_FREEBSD4) || defined(COMPAT_FREEBSD5) || \
852     defined(COMPAT_FREEBSD6) || defined(COMPAT_FREEBSD7)
853 	SYSCALL_INIT_HELPER_COMPAT(freebsd7_shmctl),
854 #endif
855 #if defined(__i386__) && (defined(COMPAT_FREEBSD4) || defined(COMPAT_43))
856 	SYSCALL_INIT_HELPER(shmsys),
857 #endif
858 	SYSCALL_INIT_LAST
859 };
860 
861 #ifdef COMPAT_FREEBSD32
862 #include <compat/freebsd32/freebsd32.h>
863 #include <compat/freebsd32/freebsd32_ipc.h>
864 #include <compat/freebsd32/freebsd32_proto.h>
865 #include <compat/freebsd32/freebsd32_signal.h>
866 #include <compat/freebsd32/freebsd32_syscall.h>
867 #include <compat/freebsd32/freebsd32_util.h>
868 
869 static struct syscall_helper_data shm32_syscalls[] = {
870 	SYSCALL32_INIT_HELPER_COMPAT(shmat),
871 	SYSCALL32_INIT_HELPER_COMPAT(shmdt),
872 	SYSCALL32_INIT_HELPER_COMPAT(shmget),
873 	SYSCALL32_INIT_HELPER(freebsd32_shmsys),
874 	SYSCALL32_INIT_HELPER(freebsd32_shmctl),
875 #if defined(COMPAT_FREEBSD4) || defined(COMPAT_FREEBSD5) || \
876     defined(COMPAT_FREEBSD6) || defined(COMPAT_FREEBSD7)
877 	SYSCALL32_INIT_HELPER(freebsd7_freebsd32_shmctl),
878 #endif
879 	SYSCALL_INIT_LAST
880 };
881 #endif
882 
883 static int
884 shminit()
885 {
886 	int i, error;
887 
888 #ifndef BURN_BRIDGES
889 	if (TUNABLE_ULONG_FETCH("kern.ipc.shmmaxpgs", &shminfo.shmall) != 0)
890 		printf("kern.ipc.shmmaxpgs is now called kern.ipc.shmall!\n");
891 #endif
892 	TUNABLE_ULONG_FETCH("kern.ipc.shmall", &shminfo.shmall);
893 	if (!TUNABLE_ULONG_FETCH("kern.ipc.shmmax", &shminfo.shmmax)) {
894 		/* Initialize shmmax dealing with possible overflow. */
895 		for (i = PAGE_SIZE; i > 0; i--) {
896 			shminfo.shmmax = shminfo.shmall * i;
897 			if (shminfo.shmmax >= shminfo.shmall)
898 				break;
899 		}
900 	}
901 	TUNABLE_ULONG_FETCH("kern.ipc.shmmin", &shminfo.shmmin);
902 	TUNABLE_ULONG_FETCH("kern.ipc.shmmni", &shminfo.shmmni);
903 	TUNABLE_ULONG_FETCH("kern.ipc.shmseg", &shminfo.shmseg);
904 	TUNABLE_INT_FETCH("kern.ipc.shm_use_phys", &shm_use_phys);
905 
906 	shmalloced = shminfo.shmmni;
907 	shmsegs = malloc(shmalloced * sizeof(shmsegs[0]), M_SHM, M_WAITOK);
908 	for (i = 0; i < shmalloced; i++) {
909 		shmsegs[i].u.shm_perm.mode = SHMSEG_FREE;
910 		shmsegs[i].u.shm_perm.seq = 0;
911 #ifdef MAC
912 		mac_sysvshm_init(&shmsegs[i]);
913 #endif
914 	}
915 	shm_last_free = 0;
916 	shm_nused = 0;
917 	shm_committed = 0;
918 	shmexit_hook = &shmexit_myhook;
919 	shmfork_hook = &shmfork_myhook;
920 
921 	error = syscall_helper_register(shm_syscalls);
922 	if (error != 0)
923 		return (error);
924 #ifdef COMPAT_FREEBSD32
925 	error = syscall32_helper_register(shm32_syscalls);
926 	if (error != 0)
927 		return (error);
928 #endif
929 	return (0);
930 }
931 
932 static int
933 shmunload()
934 {
935 	int i;
936 
937 	if (shm_nused > 0)
938 		return (EBUSY);
939 
940 #ifdef COMPAT_FREEBSD32
941 	syscall32_helper_unregister(shm32_syscalls);
942 #endif
943 	syscall_helper_unregister(shm_syscalls);
944 
945 	for (i = 0; i < shmalloced; i++) {
946 #ifdef MAC
947 		mac_sysvshm_destroy(&shmsegs[i]);
948 #endif
949 		/*
950 		 * Objects might be still mapped into the processes
951 		 * address spaces.  Actual free would happen on the
952 		 * last mapping destruction.
953 		 */
954 		if (shmsegs[i].u.shm_perm.mode != SHMSEG_FREE)
955 			vm_object_deallocate(shmsegs[i].object);
956 	}
957 	free(shmsegs, M_SHM);
958 	shmexit_hook = NULL;
959 	shmfork_hook = NULL;
960 	return (0);
961 }
962 
963 static int
964 sysctl_shmsegs(SYSCTL_HANDLER_ARGS)
965 {
966 
967 	return (SYSCTL_OUT(req, shmsegs, shmalloced * sizeof(shmsegs[0])));
968 }
969 
970 #if defined(__i386__) && (defined(COMPAT_FREEBSD4) || defined(COMPAT_43))
971 struct oshmid_ds {
972 	struct	ipc_perm_old shm_perm;	/* operation perms */
973 	int	shm_segsz;		/* size of segment (bytes) */
974 	u_short	shm_cpid;		/* pid, creator */
975 	u_short	shm_lpid;		/* pid, last operation */
976 	short	shm_nattch;		/* no. of current attaches */
977 	time_t	shm_atime;		/* last attach time */
978 	time_t	shm_dtime;		/* last detach time */
979 	time_t	shm_ctime;		/* last change time */
980 	void	*shm_handle;		/* internal handle for shm segment */
981 };
982 
983 struct oshmctl_args {
984 	int shmid;
985 	int cmd;
986 	struct oshmid_ds *ubuf;
987 };
988 
989 static int
990 oshmctl(struct thread *td, struct oshmctl_args *uap)
991 {
992 #ifdef COMPAT_43
993 	int error = 0;
994 	struct shmid_kernel *shmseg;
995 	struct oshmid_ds outbuf;
996 
997 	if (!prison_allow(td->td_ucred, PR_ALLOW_SYSVIPC))
998 		return (ENOSYS);
999 	mtx_lock(&Giant);
1000 	shmseg = shm_find_segment_by_shmid(uap->shmid);
1001 	if (shmseg == NULL) {
1002 		error = EINVAL;
1003 		goto done2;
1004 	}
1005 	switch (uap->cmd) {
1006 	case IPC_STAT:
1007 		error = ipcperm(td, &shmseg->u.shm_perm, IPC_R);
1008 		if (error)
1009 			goto done2;
1010 #ifdef MAC
1011 		error = mac_sysvshm_check_shmctl(td->td_ucred, shmseg, uap->cmd);
1012 		if (error != 0)
1013 			goto done2;
1014 #endif
1015 		ipcperm_new2old(&shmseg->u.shm_perm, &outbuf.shm_perm);
1016 		outbuf.shm_segsz = shmseg->u.shm_segsz;
1017 		outbuf.shm_cpid = shmseg->u.shm_cpid;
1018 		outbuf.shm_lpid = shmseg->u.shm_lpid;
1019 		outbuf.shm_nattch = shmseg->u.shm_nattch;
1020 		outbuf.shm_atime = shmseg->u.shm_atime;
1021 		outbuf.shm_dtime = shmseg->u.shm_dtime;
1022 		outbuf.shm_ctime = shmseg->u.shm_ctime;
1023 		outbuf.shm_handle = shmseg->object;
1024 		error = copyout(&outbuf, uap->ubuf, sizeof(outbuf));
1025 		if (error)
1026 			goto done2;
1027 		break;
1028 	default:
1029 		error = freebsd7_shmctl(td, (struct freebsd7_shmctl_args *)uap);
1030 		break;
1031 	}
1032 done2:
1033 	mtx_unlock(&Giant);
1034 	return (error);
1035 #else
1036 	return (EINVAL);
1037 #endif
1038 }
1039 
1040 /* XXX casting to (sy_call_t *) is bogus, as usual. */
1041 static sy_call_t *shmcalls[] = {
1042 	(sy_call_t *)sys_shmat, (sy_call_t *)oshmctl,
1043 	(sy_call_t *)sys_shmdt, (sy_call_t *)sys_shmget,
1044 	(sy_call_t *)freebsd7_shmctl
1045 };
1046 
1047 int
1048 sys_shmsys(td, uap)
1049 	struct thread *td;
1050 	/* XXX actually varargs. */
1051 	struct shmsys_args /* {
1052 		int	which;
1053 		int	a2;
1054 		int	a3;
1055 		int	a4;
1056 	} */ *uap;
1057 {
1058 	int error;
1059 
1060 	if (!prison_allow(td->td_ucred, PR_ALLOW_SYSVIPC))
1061 		return (ENOSYS);
1062 	if (uap->which < 0 ||
1063 	    uap->which >= sizeof(shmcalls)/sizeof(shmcalls[0]))
1064 		return (EINVAL);
1065 	mtx_lock(&Giant);
1066 	error = (*shmcalls[uap->which])(td, &uap->a2);
1067 	mtx_unlock(&Giant);
1068 	return (error);
1069 }
1070 
1071 #endif	/* i386 && (COMPAT_FREEBSD4 || COMPAT_43) */
1072 
1073 #ifdef COMPAT_FREEBSD32
1074 
1075 int
1076 freebsd32_shmsys(struct thread *td, struct freebsd32_shmsys_args *uap)
1077 {
1078 
1079 #if defined(COMPAT_FREEBSD4) || defined(COMPAT_FREEBSD5) || \
1080     defined(COMPAT_FREEBSD6) || defined(COMPAT_FREEBSD7)
1081 	switch (uap->which) {
1082 	case 0:	{	/* shmat */
1083 		struct shmat_args ap;
1084 
1085 		ap.shmid = uap->a2;
1086 		ap.shmaddr = PTRIN(uap->a3);
1087 		ap.shmflg = uap->a4;
1088 		return (sysent[SYS_shmat].sy_call(td, &ap));
1089 	}
1090 	case 2: {	/* shmdt */
1091 		struct shmdt_args ap;
1092 
1093 		ap.shmaddr = PTRIN(uap->a2);
1094 		return (sysent[SYS_shmdt].sy_call(td, &ap));
1095 	}
1096 	case 3: {	/* shmget */
1097 		struct shmget_args ap;
1098 
1099 		ap.key = uap->a2;
1100 		ap.size = uap->a3;
1101 		ap.shmflg = uap->a4;
1102 		return (sysent[SYS_shmget].sy_call(td, &ap));
1103 	}
1104 	case 4: {	/* shmctl */
1105 		struct freebsd7_freebsd32_shmctl_args ap;
1106 
1107 		ap.shmid = uap->a2;
1108 		ap.cmd = uap->a3;
1109 		ap.buf = PTRIN(uap->a4);
1110 		return (freebsd7_freebsd32_shmctl(td, &ap));
1111 	}
1112 	case 1:		/* oshmctl */
1113 	default:
1114 		return (EINVAL);
1115 	}
1116 #else
1117 	return (nosys(td, NULL));
1118 #endif
1119 }
1120 
1121 #if defined(COMPAT_FREEBSD4) || defined(COMPAT_FREEBSD5) || \
1122     defined(COMPAT_FREEBSD6) || defined(COMPAT_FREEBSD7)
1123 int
1124 freebsd7_freebsd32_shmctl(struct thread *td,
1125     struct freebsd7_freebsd32_shmctl_args *uap)
1126 {
1127 	int error = 0;
1128 	union {
1129 		struct shmid_ds shmid_ds;
1130 		struct shm_info shm_info;
1131 		struct shminfo shminfo;
1132 	} u;
1133 	union {
1134 		struct shmid_ds32_old shmid_ds32;
1135 		struct shm_info32 shm_info32;
1136 		struct shminfo32 shminfo32;
1137 	} u32;
1138 	size_t sz;
1139 
1140 	if (uap->cmd == IPC_SET) {
1141 		if ((error = copyin(uap->buf, &u32.shmid_ds32,
1142 		    sizeof(u32.shmid_ds32))))
1143 			goto done;
1144 		freebsd32_ipcperm_old_in(&u32.shmid_ds32.shm_perm,
1145 		    &u.shmid_ds.shm_perm);
1146 		CP(u32.shmid_ds32, u.shmid_ds, shm_segsz);
1147 		CP(u32.shmid_ds32, u.shmid_ds, shm_lpid);
1148 		CP(u32.shmid_ds32, u.shmid_ds, shm_cpid);
1149 		CP(u32.shmid_ds32, u.shmid_ds, shm_nattch);
1150 		CP(u32.shmid_ds32, u.shmid_ds, shm_atime);
1151 		CP(u32.shmid_ds32, u.shmid_ds, shm_dtime);
1152 		CP(u32.shmid_ds32, u.shmid_ds, shm_ctime);
1153 	}
1154 
1155 	error = kern_shmctl(td, uap->shmid, uap->cmd, (void *)&u, &sz);
1156 	if (error)
1157 		goto done;
1158 
1159 	/* Cases in which we need to copyout */
1160 	switch (uap->cmd) {
1161 	case IPC_INFO:
1162 		CP(u.shminfo, u32.shminfo32, shmmax);
1163 		CP(u.shminfo, u32.shminfo32, shmmin);
1164 		CP(u.shminfo, u32.shminfo32, shmmni);
1165 		CP(u.shminfo, u32.shminfo32, shmseg);
1166 		CP(u.shminfo, u32.shminfo32, shmall);
1167 		error = copyout(&u32.shminfo32, uap->buf,
1168 		    sizeof(u32.shminfo32));
1169 		break;
1170 	case SHM_INFO:
1171 		CP(u.shm_info, u32.shm_info32, used_ids);
1172 		CP(u.shm_info, u32.shm_info32, shm_rss);
1173 		CP(u.shm_info, u32.shm_info32, shm_tot);
1174 		CP(u.shm_info, u32.shm_info32, shm_swp);
1175 		CP(u.shm_info, u32.shm_info32, swap_attempts);
1176 		CP(u.shm_info, u32.shm_info32, swap_successes);
1177 		error = copyout(&u32.shm_info32, uap->buf,
1178 		    sizeof(u32.shm_info32));
1179 		break;
1180 	case SHM_STAT:
1181 	case IPC_STAT:
1182 		freebsd32_ipcperm_old_out(&u.shmid_ds.shm_perm,
1183 		    &u32.shmid_ds32.shm_perm);
1184 		if (u.shmid_ds.shm_segsz > INT32_MAX)
1185 			u32.shmid_ds32.shm_segsz = INT32_MAX;
1186 		else
1187 			CP(u.shmid_ds, u32.shmid_ds32, shm_segsz);
1188 		CP(u.shmid_ds, u32.shmid_ds32, shm_lpid);
1189 		CP(u.shmid_ds, u32.shmid_ds32, shm_cpid);
1190 		CP(u.shmid_ds, u32.shmid_ds32, shm_nattch);
1191 		CP(u.shmid_ds, u32.shmid_ds32, shm_atime);
1192 		CP(u.shmid_ds, u32.shmid_ds32, shm_dtime);
1193 		CP(u.shmid_ds, u32.shmid_ds32, shm_ctime);
1194 		u32.shmid_ds32.shm_internal = 0;
1195 		error = copyout(&u32.shmid_ds32, uap->buf,
1196 		    sizeof(u32.shmid_ds32));
1197 		break;
1198 	}
1199 
1200 done:
1201 	if (error) {
1202 		/* Invalidate the return value */
1203 		td->td_retval[0] = -1;
1204 	}
1205 	return (error);
1206 }
1207 #endif
1208 
1209 int
1210 freebsd32_shmctl(struct thread *td, struct freebsd32_shmctl_args *uap)
1211 {
1212 	int error = 0;
1213 	union {
1214 		struct shmid_ds shmid_ds;
1215 		struct shm_info shm_info;
1216 		struct shminfo shminfo;
1217 	} u;
1218 	union {
1219 		struct shmid_ds32 shmid_ds32;
1220 		struct shm_info32 shm_info32;
1221 		struct shminfo32 shminfo32;
1222 	} u32;
1223 	size_t sz;
1224 
1225 	if (uap->cmd == IPC_SET) {
1226 		if ((error = copyin(uap->buf, &u32.shmid_ds32,
1227 		    sizeof(u32.shmid_ds32))))
1228 			goto done;
1229 		freebsd32_ipcperm_in(&u32.shmid_ds32.shm_perm,
1230 		    &u.shmid_ds.shm_perm);
1231 		CP(u32.shmid_ds32, u.shmid_ds, shm_segsz);
1232 		CP(u32.shmid_ds32, u.shmid_ds, shm_lpid);
1233 		CP(u32.shmid_ds32, u.shmid_ds, shm_cpid);
1234 		CP(u32.shmid_ds32, u.shmid_ds, shm_nattch);
1235 		CP(u32.shmid_ds32, u.shmid_ds, shm_atime);
1236 		CP(u32.shmid_ds32, u.shmid_ds, shm_dtime);
1237 		CP(u32.shmid_ds32, u.shmid_ds, shm_ctime);
1238 	}
1239 
1240 	error = kern_shmctl(td, uap->shmid, uap->cmd, (void *)&u, &sz);
1241 	if (error)
1242 		goto done;
1243 
1244 	/* Cases in which we need to copyout */
1245 	switch (uap->cmd) {
1246 	case IPC_INFO:
1247 		CP(u.shminfo, u32.shminfo32, shmmax);
1248 		CP(u.shminfo, u32.shminfo32, shmmin);
1249 		CP(u.shminfo, u32.shminfo32, shmmni);
1250 		CP(u.shminfo, u32.shminfo32, shmseg);
1251 		CP(u.shminfo, u32.shminfo32, shmall);
1252 		error = copyout(&u32.shminfo32, uap->buf,
1253 		    sizeof(u32.shminfo32));
1254 		break;
1255 	case SHM_INFO:
1256 		CP(u.shm_info, u32.shm_info32, used_ids);
1257 		CP(u.shm_info, u32.shm_info32, shm_rss);
1258 		CP(u.shm_info, u32.shm_info32, shm_tot);
1259 		CP(u.shm_info, u32.shm_info32, shm_swp);
1260 		CP(u.shm_info, u32.shm_info32, swap_attempts);
1261 		CP(u.shm_info, u32.shm_info32, swap_successes);
1262 		error = copyout(&u32.shm_info32, uap->buf,
1263 		    sizeof(u32.shm_info32));
1264 		break;
1265 	case SHM_STAT:
1266 	case IPC_STAT:
1267 		freebsd32_ipcperm_out(&u.shmid_ds.shm_perm,
1268 		    &u32.shmid_ds32.shm_perm);
1269 		if (u.shmid_ds.shm_segsz > INT32_MAX)
1270 			u32.shmid_ds32.shm_segsz = INT32_MAX;
1271 		else
1272 			CP(u.shmid_ds, u32.shmid_ds32, shm_segsz);
1273 		CP(u.shmid_ds, u32.shmid_ds32, shm_lpid);
1274 		CP(u.shmid_ds, u32.shmid_ds32, shm_cpid);
1275 		CP(u.shmid_ds, u32.shmid_ds32, shm_nattch);
1276 		CP(u.shmid_ds, u32.shmid_ds32, shm_atime);
1277 		CP(u.shmid_ds, u32.shmid_ds32, shm_dtime);
1278 		CP(u.shmid_ds, u32.shmid_ds32, shm_ctime);
1279 		error = copyout(&u32.shmid_ds32, uap->buf,
1280 		    sizeof(u32.shmid_ds32));
1281 		break;
1282 	}
1283 
1284 done:
1285 	if (error) {
1286 		/* Invalidate the return value */
1287 		td->td_retval[0] = -1;
1288 	}
1289 	return (error);
1290 }
1291 #endif
1292 
1293 #if defined(COMPAT_FREEBSD4) || defined(COMPAT_FREEBSD5) || \
1294     defined(COMPAT_FREEBSD6) || defined(COMPAT_FREEBSD7)
1295 
1296 #ifndef CP
1297 #define CP(src, dst, fld)	do { (dst).fld = (src).fld; } while (0)
1298 #endif
1299 
1300 #ifndef _SYS_SYSPROTO_H_
1301 struct freebsd7_shmctl_args {
1302 	int shmid;
1303 	int cmd;
1304 	struct shmid_ds_old *buf;
1305 };
1306 #endif
1307 int
1308 freebsd7_shmctl(td, uap)
1309 	struct thread *td;
1310 	struct freebsd7_shmctl_args *uap;
1311 {
1312 	int error = 0;
1313 	struct shmid_ds_old old;
1314 	struct shmid_ds buf;
1315 	size_t bufsz;
1316 
1317 	/*
1318 	 * The only reason IPC_INFO, SHM_INFO, SHM_STAT exists is to support
1319 	 * Linux binaries.  If we see the call come through the FreeBSD ABI,
1320 	 * return an error back to the user since we do not to support this.
1321 	 */
1322 	if (uap->cmd == IPC_INFO || uap->cmd == SHM_INFO ||
1323 	    uap->cmd == SHM_STAT)
1324 		return (EINVAL);
1325 
1326 	/* IPC_SET needs to copyin the buffer before calling kern_shmctl */
1327 	if (uap->cmd == IPC_SET) {
1328 		if ((error = copyin(uap->buf, &old, sizeof(old))))
1329 			goto done;
1330 		ipcperm_old2new(&old.shm_perm, &buf.shm_perm);
1331 		CP(old, buf, shm_segsz);
1332 		CP(old, buf, shm_lpid);
1333 		CP(old, buf, shm_cpid);
1334 		CP(old, buf, shm_nattch);
1335 		CP(old, buf, shm_atime);
1336 		CP(old, buf, shm_dtime);
1337 		CP(old, buf, shm_ctime);
1338 	}
1339 
1340 	error = kern_shmctl(td, uap->shmid, uap->cmd, (void *)&buf, &bufsz);
1341 	if (error)
1342 		goto done;
1343 
1344 	/* Cases in which we need to copyout */
1345 	switch (uap->cmd) {
1346 	case IPC_STAT:
1347 		ipcperm_new2old(&buf.shm_perm, &old.shm_perm);
1348 		if (buf.shm_segsz > INT_MAX)
1349 			old.shm_segsz = INT_MAX;
1350 		else
1351 			CP(buf, old, shm_segsz);
1352 		CP(buf, old, shm_lpid);
1353 		CP(buf, old, shm_cpid);
1354 		if (buf.shm_nattch > SHRT_MAX)
1355 			old.shm_nattch = SHRT_MAX;
1356 		else
1357 			CP(buf, old, shm_nattch);
1358 		CP(buf, old, shm_atime);
1359 		CP(buf, old, shm_dtime);
1360 		CP(buf, old, shm_ctime);
1361 		old.shm_internal = NULL;
1362 		error = copyout(&old, uap->buf, sizeof(old));
1363 		break;
1364 	}
1365 
1366 done:
1367 	if (error) {
1368 		/* Invalidate the return value */
1369 		td->td_retval[0] = -1;
1370 	}
1371 	return (error);
1372 }
1373 
1374 #endif	/* COMPAT_FREEBSD4 || COMPAT_FREEBSD5 || COMPAT_FREEBSD6 ||
1375 	   COMPAT_FREEBSD7 */
1376 
1377 static int
1378 sysvshm_modload(struct module *module, int cmd, void *arg)
1379 {
1380 	int error = 0;
1381 
1382 	switch (cmd) {
1383 	case MOD_LOAD:
1384 		error = shminit();
1385 		if (error != 0)
1386 			shmunload();
1387 		break;
1388 	case MOD_UNLOAD:
1389 		error = shmunload();
1390 		break;
1391 	case MOD_SHUTDOWN:
1392 		break;
1393 	default:
1394 		error = EINVAL;
1395 		break;
1396 	}
1397 	return (error);
1398 }
1399 
1400 static moduledata_t sysvshm_mod = {
1401 	"sysvshm",
1402 	&sysvshm_modload,
1403 	NULL
1404 };
1405 
1406 DECLARE_MODULE(sysvshm, sysvshm_mod, SI_SUB_SYSV_SHM, SI_ORDER_FIRST);
1407 MODULE_VERSION(sysvshm, 1);
1408