sys/kern/sysv_shm.c

   1 /*-
   2  * SPDX-License-Identifier: BSD-4-Clause AND BSD-2-Clause-FreeBSD
   3  *
   4  * Copyright (c) 1994 Adam Glass and Charles Hannum.  All rights reserved.
   5  *
   6  * Redistribution and use in source and binary forms, with or without
   7  * modification, are permitted provided that the following conditions
   8  * are met:
   9  * 1. Redistributions of source code must retain the above copyright
  10  *    notice, this list of conditions and the following disclaimer.
  11  * 2. Redistributions in binary form must reproduce the above copyright
  12  *    notice, this list of conditions and the following disclaimer in the
  13  *    documentation and/or other materials provided with the distribution.
  14  * 3. All advertising materials mentioning features or use of this software
  15  *    must display the following acknowledgement:
  16  *      This product includes software developed by Adam Glass and Charles
  17  *      Hannum.
  18  * 4. The names of the authors may not be used to endorse or promote products
  19  *    derived from this software without specific prior written permission.
  20  *
  21  * THIS SOFTWARE IS PROVIDED BY THE AUTHORS ``AS IS'' AND ANY EXPRESS OR
  22  * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
  23  * OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
  24  * IN NO EVENT SHALL THE AUTHORS BE LIABLE FOR ANY DIRECT, INDIRECT,
  25  * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
  26  * NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
  27  * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
  28  * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
  29  * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
  30  * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
  31  *
  32  * $NetBSD: sysv_shm.c,v 1.39 1997/10/07 10:02:03 drochner Exp $
  33  */
  34 /*-
  35  * Copyright (c) 2003-2005 McAfee, Inc.
  36  * Copyright (c) 2016-2017 Robert N. M. Watson
  37  * All rights reserved.
  38  *
  39  * This software was developed for the FreeBSD Project in part by McAfee
  40  * Research, the Security Research Division of McAfee, Inc under DARPA/SPAWAR
  41  * contract N66001-01-C-8035 ("CBOSS"), as part of the DARPA CHATS research
  42  * program.
  43  *
  44  * Portions of this software were developed by BAE Systems, the University of
  45  * Cambridge Computer Laboratory, and Memorial University under DARPA/AFRL
  46  * contract FA8650-15-C-7558 ("CADETS"), as part of the DARPA Transparent
  47  * Computing (TC) research program.
  48  *
  49  * Redistribution and use in source and binary forms, with or without
  50  * modification, are permitted provided that the following conditions
  51  * are met:
  52  * 1. Redistributions of source code must retain the above copyright
  53  *    notice, this list of conditions and the following disclaimer.
  54  * 2. Redistributions in binary form must reproduce the above copyright
  55  *    notice, this list of conditions and the following disclaimer in the
  56  *    documentation and/or other materials provided with the distribution.
  57  *
  58  * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
  59  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
  60  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
  61  * ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
  62  * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
  63  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
  64  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
  65  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
  66  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
  67  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
  68  * SUCH DAMAGE.
  69  */
  70
  71 #include <sys/cdefs.h>
  72 __FBSDID("$FreeBSD$");
  73
  74 #include "opt_compat.h"
  75 #include "opt_sysvipc.h"
  76
  77 #include <sys/param.h>
  78 #include <sys/systm.h>
  79 #include <sys/kernel.h>
  80 #include <sys/limits.h>
  81 #include <sys/lock.h>
  82 #include <sys/sysctl.h>
  83 #include <sys/shm.h>
  84 #include <sys/proc.h>
  85 #include <sys/malloc.h>
  86 #include <sys/mman.h>
  87 #include <sys/module.h>
  88 #include <sys/mutex.h>
  89 #include <sys/racct.h>
  90 #include <sys/resourcevar.h>
  91 #include <sys/rwlock.h>
  92 #include <sys/stat.h>
  93 #include <sys/syscall.h>
  94 #include <sys/syscallsubr.h>
  95 #include <sys/sysent.h>
  96 #include <sys/sysproto.h>
  97 #include <sys/jail.h>
  98
  99 #include <security/audit/audit.h>
 100 #include <security/mac/mac_framework.h>
 101
 102 #include <vm/vm.h>
 103 #include <vm/vm_param.h>
 104 #include <vm/pmap.h>
 105 #include <vm/vm_object.h>
 106 #include <vm/vm_map.h>
 107 #include <vm/vm_page.h>
 108 #include <vm/vm_pager.h>
 109
 110 FEATURE(sysv_shm, "System V shared memory segments support");
 111
 112 static MALLOC_DEFINE(M_SHM, "shm", "SVID compatible shared memory segments");
 113
 114 static int shmget_allocate_segment(struct thread *td,
 115     struct shmget_args *uap, int mode);
 116 static int shmget_existing(struct thread *td, struct shmget_args *uap,
 117     int mode, int segnum);
 118
 119 #define SHMSEG_FREE             0x0200
 120 #define SHMSEG_REMOVED          0x0400
 121 #define SHMSEG_ALLOCATED        0x0800
 122
 123 static int shm_last_free, shm_nused, shmalloced;
 124 vm_size_t shm_committed;
 125 static struct shmid_kernel *shmsegs;
 126 static unsigned shm_prison_slot;
 127
 128 struct shmmap_state {
 129         vm_offset_t va;
 130         int shmid;
 131 };
 132
 133 static void shm_deallocate_segment(struct shmid_kernel *);
 134 static int shm_find_segment_by_key(struct prison *, key_t);
 135 static struct shmid_kernel *shm_find_segment(struct prison *, int, bool);
 136 static int shm_delete_mapping(struct vmspace *vm, struct shmmap_state *);
 137 static void shmrealloc(void);
 138 static int shminit(void);
 139 static int sysvshm_modload(struct module *, int, void *);
 140 static int shmunload(void);
 141 static void shmexit_myhook(struct vmspace *vm);
 142 static void shmfork_myhook(struct proc *p1, struct proc *p2);
 143 static int sysctl_shmsegs(SYSCTL_HANDLER_ARGS);
 144 static void shm_remove(struct shmid_kernel *, int);
 145 static struct prison *shm_find_prison(struct ucred *);
 146 static int shm_prison_cansee(struct prison *, struct shmid_kernel *);
 147 static int shm_prison_check(void *, void *);
 148 static int shm_prison_set(void *, void *);
 149 static int shm_prison_get(void *, void *);
 150 static int shm_prison_remove(void *, void *);
 151 static void shm_prison_cleanup(struct prison *);
 152
 153 /*
 154  * Tuneable values.
 155  */
 156 #ifndef SHMMAXPGS
 157 #define SHMMAXPGS       131072  /* Note: sysv shared memory is swap backed. */
 158 #endif
 159 #ifndef SHMMAX
 160 #define SHMMAX  (SHMMAXPGS*PAGE_SIZE)
 161 #endif
 162 #ifndef SHMMIN
 163 #define SHMMIN  1
 164 #endif
 165 #ifndef SHMMNI
 166 #define SHMMNI  192
 167 #endif
 168 #ifndef SHMSEG
 169 #define SHMSEG  128
 170 #endif
 171 #ifndef SHMALL
 172 #define SHMALL  (SHMMAXPGS)
 173 #endif
 174
 175 struct  shminfo shminfo = {
 176         .shmmax = SHMMAX,
 177         .shmmin = SHMMIN,
 178         .shmmni = SHMMNI,
 179         .shmseg = SHMSEG,
 180         .shmall = SHMALL
 181 };
 182
 183 static int shm_use_phys;
 184 static int shm_allow_removed = 1;
 185
 186 SYSCTL_ULONG(_kern_ipc, OID_AUTO, shmmax, CTLFLAG_RWTUN, &shminfo.shmmax, 0,
 187     "Maximum shared memory segment size");
 188 SYSCTL_ULONG(_kern_ipc, OID_AUTO, shmmin, CTLFLAG_RWTUN, &shminfo.shmmin, 0,
 189     "Minimum shared memory segment size");
 190 SYSCTL_ULONG(_kern_ipc, OID_AUTO, shmmni, CTLFLAG_RDTUN, &shminfo.shmmni, 0,
 191     "Number of shared memory identifiers");
 192 SYSCTL_ULONG(_kern_ipc, OID_AUTO, shmseg, CTLFLAG_RDTUN, &shminfo.shmseg, 0,
 193     "Number of segments per process");
 194 SYSCTL_ULONG(_kern_ipc, OID_AUTO, shmall, CTLFLAG_RWTUN, &shminfo.shmall, 0,
 195     "Maximum number of pages available for shared memory");
 196 SYSCTL_INT(_kern_ipc, OID_AUTO, shm_use_phys, CTLFLAG_RWTUN,
 197     &shm_use_phys, 0, "Enable/Disable locking of shared memory pages in core");
 198 SYSCTL_INT(_kern_ipc, OID_AUTO, shm_allow_removed, CTLFLAG_RWTUN,
 199     &shm_allow_removed, 0,
 200     "Enable/Disable attachment to attached segments marked for removal");
 201 SYSCTL_PROC(_kern_ipc, OID_AUTO, shmsegs, CTLTYPE_OPAQUE | CTLFLAG_RD |
 202     CTLFLAG_MPSAFE, NULL, 0, sysctl_shmsegs, "",
 203     "Array of struct shmid_kernel for each potential shared memory segment");
 204
 205 static struct sx sysvshmsx;
 206 #define SYSVSHM_LOCK()          sx_xlock(&sysvshmsx)
 207 #define SYSVSHM_UNLOCK()        sx_xunlock(&sysvshmsx)
 208 #define SYSVSHM_ASSERT_LOCKED() sx_assert(&sysvshmsx, SA_XLOCKED)
 209
 210 static int
 211 shm_find_segment_by_key(struct prison *pr, key_t key)
 212 {
 213         int i;
 214
 215         for (i = 0; i < shmalloced; i++)
 216                 if ((shmsegs[i].u.shm_perm.mode & SHMSEG_ALLOCATED) &&
 217                     shmsegs[i].cred != NULL &&
 218                     shmsegs[i].cred->cr_prison == pr &&
 219                     shmsegs[i].u.shm_perm.key == key)
 220                         return (i);
 221         return (-1);
 222 }
 223
 224 /*
 225  * Finds segment either by shmid if is_shmid is true, or by segnum if
 226  * is_shmid is false.
 227  */
 228 static struct shmid_kernel *
 229 shm_find_segment(struct prison *rpr, int arg, bool is_shmid)
 230 {
 231         struct shmid_kernel *shmseg;
 232         int segnum;
 233
 234         segnum = is_shmid ? IPCID_TO_IX(arg) : arg;
 235         if (segnum < 0 || segnum >= shmalloced)
 236                 return (NULL);
 237         shmseg = &shmsegs[segnum];
 238         if ((shmseg->u.shm_perm.mode & SHMSEG_ALLOCATED) == 0 ||
 239             (!shm_allow_removed &&
 240             (shmseg->u.shm_perm.mode & SHMSEG_REMOVED) != 0) ||
 241             (is_shmid && shmseg->u.shm_perm.seq != IPCID_TO_SEQ(arg)) ||
 242             shm_prison_cansee(rpr, shmseg) != 0)
 243                 return (NULL);
 244         return (shmseg);
 245 }
 246
 247 static void
 248 shm_deallocate_segment(struct shmid_kernel *shmseg)
 249 {
 250         vm_size_t size;
 251
 252         SYSVSHM_ASSERT_LOCKED();
 253
 254         vm_object_deallocate(shmseg->object);
 255         shmseg->object = NULL;
 256         size = round_page(shmseg->u.shm_segsz);
 257         shm_committed -= btoc(size);
 258         shm_nused--;
 259         shmseg->u.shm_perm.mode = SHMSEG_FREE;
 260 #ifdef MAC
 261         mac_sysvshm_cleanup(shmseg);
 262 #endif
 263         racct_sub_cred(shmseg->cred, RACCT_NSHM, 1);
 264         racct_sub_cred(shmseg->cred, RACCT_SHMSIZE, size);
 265         crfree(shmseg->cred);
 266         shmseg->cred = NULL;
 267 }
 268
 269 static int
 270 shm_delete_mapping(struct vmspace *vm, struct shmmap_state *shmmap_s)
 271 {
 272         struct shmid_kernel *shmseg;
 273         int segnum, result;
 274         vm_size_t size;
 275
 276         SYSVSHM_ASSERT_LOCKED();
 277         segnum = IPCID_TO_IX(shmmap_s->shmid);
 278         KASSERT(segnum >= 0 && segnum < shmalloced,
 279             ("segnum %d shmalloced %d", segnum, shmalloced));
 280
 281         shmseg = &shmsegs[segnum];
 282         size = round_page(shmseg->u.shm_segsz);
 283         result = vm_map_remove(&vm->vm_map, shmmap_s->va, shmmap_s->va + size);
 284         if (result != KERN_SUCCESS)
 285                 return (EINVAL);
 286         shmmap_s->shmid = -1;
 287         shmseg->u.shm_dtime = time_second;
 288         if (--shmseg->u.shm_nattch == 0 &&
 289             (shmseg->u.shm_perm.mode & SHMSEG_REMOVED)) {
 290                 shm_deallocate_segment(shmseg);
 291                 shm_last_free = segnum;
 292         }
 293         return (0);
 294 }
 295
 296 static void
 297 shm_remove(struct shmid_kernel *shmseg, int segnum)
 298 {
 299
 300         shmseg->u.shm_perm.key = IPC_PRIVATE;
 301         shmseg->u.shm_perm.mode |= SHMSEG_REMOVED;
 302         if (shmseg->u.shm_nattch == 0) {
 303                 shm_deallocate_segment(shmseg);
 304                 shm_last_free = segnum;
 305         }
 306 }
 307
 308 static struct prison *
 309 shm_find_prison(struct ucred *cred)
 310 {
 311         struct prison *pr, *rpr;
 312
 313         pr = cred->cr_prison;
 314         prison_lock(pr);
 315         rpr = osd_jail_get(pr, shm_prison_slot);
 316         prison_unlock(pr);
 317         return rpr;
 318 }
 319
 320 static int
 321 shm_prison_cansee(struct prison *rpr, struct shmid_kernel *shmseg)
 322 {
 323
 324         if (shmseg->cred == NULL ||
 325             !(rpr == shmseg->cred->cr_prison ||
 326               prison_ischild(rpr, shmseg->cred->cr_prison)))
 327                 return (EINVAL);
 328         return (0);
 329 }
 330
 331 static int
 332 kern_shmdt_locked(struct thread *td, const void *shmaddr)
 333 {
 334         struct proc *p = td->td_proc;
 335         struct shmmap_state *shmmap_s;
 336 #if defined(AUDIT) || defined(MAC)
 337         struct shmid_kernel *shmsegptr;
 338 #endif
 339 #ifdef MAC
 340         int error;
 341 #endif
 342         int i;
 343
 344         SYSVSHM_ASSERT_LOCKED();
 345         if (shm_find_prison(td->td_ucred) == NULL)
 346                 return (ENOSYS);
 347         shmmap_s = p->p_vmspace->vm_shm;
 348         if (shmmap_s == NULL)
 349                 return (EINVAL);
 350         AUDIT_ARG_SVIPC_ID(shmmap_s->shmid);
 351         for (i = 0; i < shminfo.shmseg; i++, shmmap_s++) {
 352                 if (shmmap_s->shmid != -1 &&
 353                     shmmap_s->va == (vm_offset_t)shmaddr) {
 354                         break;
 355                 }
 356         }
 357         if (i == shminfo.shmseg)
 358                 return (EINVAL);
 359 #if (defined(AUDIT) && defined(KDTRACE_HOOKS)) || defined(MAC)
 360         shmsegptr = &shmsegs[IPCID_TO_IX(shmmap_s->shmid)];
 361 #endif
 362 #ifdef MAC
 363         error = mac_sysvshm_check_shmdt(td->td_ucred, shmsegptr);
 364         if (error != 0)
 365                 return (error);
 366 #endif
 367         return (shm_delete_mapping(p->p_vmspace, shmmap_s));
 368 }
 369
 370 #ifndef _SYS_SYSPROTO_H_
 371 struct shmdt_args {
 372         const void *shmaddr;
 373 };
 374 #endif
 375 int
 376 sys_shmdt(struct thread *td, struct shmdt_args *uap)
 377 {
 378         int error;
 379
 380         SYSVSHM_LOCK();
 381         error = kern_shmdt_locked(td, uap->shmaddr);
 382         SYSVSHM_UNLOCK();
 383         return (error);
 384 }
 385
 386 static int
 387 kern_shmat_locked(struct thread *td, int shmid, const void *shmaddr,
 388     int shmflg)
 389 {
 390         struct prison *rpr;
 391         struct proc *p = td->td_proc;
 392         struct shmid_kernel *shmseg;
 393         struct shmmap_state *shmmap_s;
 394         vm_offset_t attach_va;
 395         vm_prot_t prot;
 396         vm_size_t size;
 397         int error, i, rv;
 398
 399         AUDIT_ARG_SVIPC_ID(shmid);
 400         AUDIT_ARG_VALUE(shmflg);
 401
 402         SYSVSHM_ASSERT_LOCKED();
 403         rpr = shm_find_prison(td->td_ucred);
 404         if (rpr == NULL)
 405                 return (ENOSYS);
 406         shmmap_s = p->p_vmspace->vm_shm;
 407         if (shmmap_s == NULL) {
 408                 shmmap_s = malloc(shminfo.shmseg * sizeof(struct shmmap_state),
 409                     M_SHM, M_WAITOK);
 410                 for (i = 0; i < shminfo.shmseg; i++)
 411                         shmmap_s[i].shmid = -1;
 412                 KASSERT(p->p_vmspace->vm_shm == NULL, ("raced"));
 413                 p->p_vmspace->vm_shm = shmmap_s;
 414         }
 415         shmseg = shm_find_segment(rpr, shmid, true);
 416         if (shmseg == NULL)
 417                 return (EINVAL);
 418         error = ipcperm(td, &shmseg->u.shm_perm,
 419             (shmflg & SHM_RDONLY) ? IPC_R : IPC_R|IPC_W);
 420         if (error != 0)
 421                 return (error);
 422 #ifdef MAC
 423         error = mac_sysvshm_check_shmat(td->td_ucred, shmseg, shmflg);
 424         if (error != 0)
 425                 return (error);
 426 #endif
 427         for (i = 0; i < shminfo.shmseg; i++) {
 428                 if (shmmap_s->shmid == -1)
 429                         break;
 430                 shmmap_s++;
 431         }
 432         if (i >= shminfo.shmseg)
 433                 return (EMFILE);
 434         size = round_page(shmseg->u.shm_segsz);
 435         prot = VM_PROT_READ;
 436         if ((shmflg & SHM_RDONLY) == 0)
 437                 prot |= VM_PROT_WRITE;
 438         if (shmaddr != NULL) {
 439                 if ((shmflg & SHM_RND) != 0)
 440                         attach_va = rounddown2((vm_offset_t)shmaddr, SHMLBA);
 441                 else if (((vm_offset_t)shmaddr & (SHMLBA-1)) == 0)
 442                         attach_va = (vm_offset_t)shmaddr;
 443                 else
 444                         return (EINVAL);
 445         } else {
 446                 /*
 447                  * This is just a hint to vm_map_find() about where to
 448                  * put it.
 449                  */
 450                 attach_va = round_page((vm_offset_t)p->p_vmspace->vm_daddr +
 451                     lim_max(td, RLIMIT_DATA));
 452         }
 453
 454         vm_object_reference(shmseg->object);
 455         rv = vm_map_find(&p->p_vmspace->vm_map, shmseg->object, 0, &attach_va,
 456             size, 0, shmaddr != NULL ? VMFS_NO_SPACE : VMFS_OPTIMAL_SPACE,
 457             prot, prot, MAP_INHERIT_SHARE | MAP_PREFAULT_PARTIAL);
 458         if (rv != KERN_SUCCESS) {
 459                 vm_object_deallocate(shmseg->object);
 460                 return (ENOMEM);
 461         }
 462
 463         shmmap_s->va = attach_va;
 464         shmmap_s->shmid = shmid;
 465         shmseg->u.shm_lpid = p->p_pid;
 466         shmseg->u.shm_atime = time_second;
 467         shmseg->u.shm_nattch++;
 468         td->td_retval[0] = attach_va;
 469         return (error);
 470 }
 471
 472 int
 473 kern_shmat(struct thread *td, int shmid, const void *shmaddr, int shmflg)
 474 {
 475         int error;
 476
 477         SYSVSHM_LOCK();
 478         error = kern_shmat_locked(td, shmid, shmaddr, shmflg);
 479         SYSVSHM_UNLOCK();
 480         return (error);
 481 }
 482
 483 #ifndef _SYS_SYSPROTO_H_
 484 struct shmat_args {
 485         int shmid;
 486         const void *shmaddr;
 487         int shmflg;
 488 };
 489 #endif
 490 int
 491 sys_shmat(struct thread *td, struct shmat_args *uap)
 492 {
 493
 494         return (kern_shmat(td, uap->shmid, uap->shmaddr, uap->shmflg));
 495 }
 496
 497 static int
 498 kern_shmctl_locked(struct thread *td, int shmid, int cmd, void *buf,
 499     size_t *bufsz)
 500 {
 501         struct prison *rpr;
 502         struct shmid_kernel *shmseg;
 503         struct shmid_ds *shmidp;
 504         struct shm_info shm_info;
 505         int error;
 506
 507         SYSVSHM_ASSERT_LOCKED();
 508
 509         rpr = shm_find_prison(td->td_ucred);
 510         if (rpr == NULL)
 511                 return (ENOSYS);
 512
 513         AUDIT_ARG_SVIPC_ID(shmid);
 514         AUDIT_ARG_SVIPC_CMD(cmd);
 515
 516         switch (cmd) {
 517         /*
 518          * It is possible that kern_shmctl is being called from the Linux ABI
 519          * layer, in which case, we will need to implement IPC_INFO.  It should
 520          * be noted that other shmctl calls will be funneled through here for
 521          * Linix binaries as well.
 522          *
 523          * NB: The Linux ABI layer will convert this data to structure(s) more
 524          * consistent with the Linux ABI.
 525          */
 526         case IPC_INFO:
 527                 memcpy(buf, &shminfo, sizeof(shminfo));
 528                 if (bufsz)
 529                         *bufsz = sizeof(shminfo);
 530                 td->td_retval[0] = shmalloced;
 531                 return (0);
 532         case SHM_INFO: {
 533                 shm_info.used_ids = shm_nused;
 534                 shm_info.shm_rss = 0;   /*XXX where to get from ? */
 535                 shm_info.shm_tot = 0;   /*XXX where to get from ? */
 536                 shm_info.shm_swp = 0;   /*XXX where to get from ? */
 537                 shm_info.swap_attempts = 0;     /*XXX where to get from ? */
 538                 shm_info.swap_successes = 0;    /*XXX where to get from ? */
 539                 memcpy(buf, &shm_info, sizeof(shm_info));
 540                 if (bufsz != NULL)
 541                         *bufsz = sizeof(shm_info);
 542                 td->td_retval[0] = shmalloced;
 543                 return (0);
 544         }
 545         }
 546         shmseg = shm_find_segment(rpr, shmid, cmd != SHM_STAT);
 547         if (shmseg == NULL)
 548                 return (EINVAL);
 549 #ifdef MAC
 550         error = mac_sysvshm_check_shmctl(td->td_ucred, shmseg, cmd);
 551         if (error != 0)
 552                 return (error);
 553 #endif
 554         switch (cmd) {
 555         case SHM_STAT:
 556         case IPC_STAT:
 557                 shmidp = (struct shmid_ds *)buf;
 558                 error = ipcperm(td, &shmseg->u.shm_perm, IPC_R);
 559                 if (error != 0)
 560                         return (error);
 561                 memcpy(shmidp, &shmseg->u, sizeof(struct shmid_ds));
 562                 if (td->td_ucred->cr_prison != shmseg->cred->cr_prison)
 563                         shmidp->shm_perm.key = IPC_PRIVATE;
 564                 if (bufsz != NULL)
 565                         *bufsz = sizeof(struct shmid_ds);
 566                 if (cmd == SHM_STAT) {
 567                         td->td_retval[0] = IXSEQ_TO_IPCID(shmid,
 568                             shmseg->u.shm_perm);
 569                 }
 570                 break;
 571         case IPC_SET:
 572                 shmidp = (struct shmid_ds *)buf;
 573                 AUDIT_ARG_SVIPC_PERM(&shmidp->shm_perm);
 574                 error = ipcperm(td, &shmseg->u.shm_perm, IPC_M);
 575                 if (error != 0)
 576                         return (error);
 577                 shmseg->u.shm_perm.uid = shmidp->shm_perm.uid;
 578                 shmseg->u.shm_perm.gid = shmidp->shm_perm.gid;
 579                 shmseg->u.shm_perm.mode =
 580                     (shmseg->u.shm_perm.mode & ~ACCESSPERMS) |
 581                     (shmidp->shm_perm.mode & ACCESSPERMS);
 582                 shmseg->u.shm_ctime = time_second;
 583                 break;
 584         case IPC_RMID:
 585                 error = ipcperm(td, &shmseg->u.shm_perm, IPC_M);
 586                 if (error != 0)
 587                         return (error);
 588                 shm_remove(shmseg, IPCID_TO_IX(shmid));
 589                 break;
 590 #if 0
 591         case SHM_LOCK:
 592         case SHM_UNLOCK:
 593 #endif
 594         default:
 595                 error = EINVAL;
 596                 break;
 597         }
 598         return (error);
 599 }
 600
 601 int
 602 kern_shmctl(struct thread *td, int shmid, int cmd, void *buf, size_t *bufsz)
 603 {
 604         int error;
 605
 606         SYSVSHM_LOCK();
 607         error = kern_shmctl_locked(td, shmid, cmd, buf, bufsz);
 608         SYSVSHM_UNLOCK();
 609         return (error);
 610 }
 611
 612
 613 #ifndef _SYS_SYSPROTO_H_
 614 struct shmctl_args {
 615         int shmid;
 616         int cmd;
 617         struct shmid_ds *buf;
 618 };
 619 #endif
 620 int
 621 sys_shmctl(struct thread *td, struct shmctl_args *uap)
 622 {
 623         int error;
 624         struct shmid_ds buf;
 625         size_t bufsz;
 626
 627         /*
 628          * The only reason IPC_INFO, SHM_INFO, SHM_STAT exists is to support
 629          * Linux binaries.  If we see the call come through the FreeBSD ABI,
 630          * return an error back to the user since we do not to support this.
 631          */
 632         if (uap->cmd == IPC_INFO || uap->cmd == SHM_INFO ||
 633             uap->cmd == SHM_STAT)
 634                 return (EINVAL);
 635
 636         /* IPC_SET needs to copyin the buffer before calling kern_shmctl */
 637         if (uap->cmd == IPC_SET) {
 638                 if ((error = copyin(uap->buf, &buf, sizeof(struct shmid_ds))))
 639                         goto done;
 640         }
 641
 642         error = kern_shmctl(td, uap->shmid, uap->cmd, (void *)&buf, &bufsz);
 643         if (error)
 644                 goto done;
 645
 646         /* Cases in which we need to copyout */
 647         switch (uap->cmd) {
 648         case IPC_STAT:
 649                 error = copyout(&buf, uap->buf, bufsz);
 650                 break;
 651         }
 652
 653 done:
 654         if (error) {
 655                 /* Invalidate the return value */
 656                 td->td_retval[0] = -1;
 657         }
 658         return (error);
 659 }
 660
 661
 662 static int
 663 shmget_existing(struct thread *td, struct shmget_args *uap, int mode,
 664     int segnum)
 665 {
 666         struct shmid_kernel *shmseg;
 667 #ifdef MAC
 668         int error;
 669 #endif
 670
 671         SYSVSHM_ASSERT_LOCKED();
 672         KASSERT(segnum >= 0 && segnum < shmalloced,
 673             ("segnum %d shmalloced %d", segnum, shmalloced));
 674         shmseg = &shmsegs[segnum];
 675         if ((uap->shmflg & (IPC_CREAT | IPC_EXCL)) == (IPC_CREAT | IPC_EXCL))
 676                 return (EEXIST);
 677 #ifdef MAC
 678         error = mac_sysvshm_check_shmget(td->td_ucred, shmseg, uap->shmflg);
 679         if (error != 0)
 680                 return (error);
 681 #endif
 682         if (uap->size != 0 && uap->size > shmseg->u.shm_segsz)
 683                 return (EINVAL);
 684         td->td_retval[0] = IXSEQ_TO_IPCID(segnum, shmseg->u.shm_perm);
 685         return (0);
 686 }
 687
 688 static int
 689 shmget_allocate_segment(struct thread *td, struct shmget_args *uap, int mode)
 690 {
 691         struct ucred *cred = td->td_ucred;
 692         struct shmid_kernel *shmseg;
 693         vm_object_t shm_object;
 694         int i, segnum;
 695         size_t size;
 696
 697         SYSVSHM_ASSERT_LOCKED();
 698
 699         if (uap->size < shminfo.shmmin || uap->size > shminfo.shmmax)
 700                 return (EINVAL);
 701         if (shm_nused >= shminfo.shmmni) /* Any shmids left? */
 702                 return (ENOSPC);
 703         size = round_page(uap->size);
 704         if (shm_committed + btoc(size) > shminfo.shmall)
 705                 return (ENOMEM);
 706         if (shm_last_free < 0) {
 707                 shmrealloc();   /* Maybe expand the shmsegs[] array. */
 708                 for (i = 0; i < shmalloced; i++)
 709                         if (shmsegs[i].u.shm_perm.mode & SHMSEG_FREE)
 710                                 break;
 711                 if (i == shmalloced)
 712                         return (ENOSPC);
 713                 segnum = i;
 714         } else  {
 715                 segnum = shm_last_free;
 716                 shm_last_free = -1;
 717         }
 718         KASSERT(segnum >= 0 && segnum < shmalloced,
 719             ("segnum %d shmalloced %d", segnum, shmalloced));
 720         shmseg = &shmsegs[segnum];
 721 #ifdef RACCT
 722         if (racct_enable) {
 723                 PROC_LOCK(td->td_proc);
 724                 if (racct_add(td->td_proc, RACCT_NSHM, 1)) {
 725                         PROC_UNLOCK(td->td_proc);
 726                         return (ENOSPC);
 727                 }
 728                 if (racct_add(td->td_proc, RACCT_SHMSIZE, size)) {
 729                         racct_sub(td->td_proc, RACCT_NSHM, 1);
 730                         PROC_UNLOCK(td->td_proc);
 731                         return (ENOMEM);
 732                 }
 733                 PROC_UNLOCK(td->td_proc);
 734         }
 735 #endif
 736
 737         /*
 738          * We make sure that we have allocated a pager before we need
 739          * to.
 740          */
 741         shm_object = vm_pager_allocate(shm_use_phys ? OBJT_PHYS : OBJT_SWAP,
 742             0, size, VM_PROT_DEFAULT, 0, cred);
 743         if (shm_object == NULL) {
 744 #ifdef RACCT
 745                 if (racct_enable) {
 746                         PROC_LOCK(td->td_proc);
 747                         racct_sub(td->td_proc, RACCT_NSHM, 1);
 748                         racct_sub(td->td_proc, RACCT_SHMSIZE, size);
 749                         PROC_UNLOCK(td->td_proc);
 750                 }
 751 #endif
 752                 return (ENOMEM);
 753         }
 754         shm_object->pg_color = 0;
 755         VM_OBJECT_WLOCK(shm_object);
 756         vm_object_clear_flag(shm_object, OBJ_ONEMAPPING);
 757         vm_object_set_flag(shm_object, OBJ_COLORED | OBJ_NOSPLIT);
 758         VM_OBJECT_WUNLOCK(shm_object);
 759
 760         shmseg->object = shm_object;
 761         shmseg->u.shm_perm.cuid = shmseg->u.shm_perm.uid = cred->cr_uid;
 762         shmseg->u.shm_perm.cgid = shmseg->u.shm_perm.gid = cred->cr_gid;
 763         shmseg->u.shm_perm.mode = (mode & ACCESSPERMS) | SHMSEG_ALLOCATED;
 764         shmseg->u.shm_perm.key = uap->key;
 765         shmseg->u.shm_perm.seq = (shmseg->u.shm_perm.seq + 1) & 0x7fff;
 766         shmseg->cred = crhold(cred);
 767         shmseg->u.shm_segsz = uap->size;
 768         shmseg->u.shm_cpid = td->td_proc->p_pid;
 769         shmseg->u.shm_lpid = shmseg->u.shm_nattch = 0;
 770         shmseg->u.shm_atime = shmseg->u.shm_dtime = 0;
 771 #ifdef MAC
 772         mac_sysvshm_create(cred, shmseg);
 773 #endif
 774         shmseg->u.shm_ctime = time_second;
 775         shm_committed += btoc(size);
 776         shm_nused++;
 777         td->td_retval[0] = IXSEQ_TO_IPCID(segnum, shmseg->u.shm_perm);
 778
 779         return (0);
 780 }
 781
 782 #ifndef _SYS_SYSPROTO_H_
 783 struct shmget_args {
 784         key_t key;
 785         size_t size;
 786         int shmflg;
 787 };
 788 #endif
 789 int
 790 sys_shmget(struct thread *td, struct shmget_args *uap)
 791 {
 792         int segnum, mode;
 793         int error;
 794
 795         if (shm_find_prison(td->td_ucred) == NULL)
 796                 return (ENOSYS);
 797         mode = uap->shmflg & ACCESSPERMS;
 798         SYSVSHM_LOCK();
 799         if (uap->key == IPC_PRIVATE) {
 800                 error = shmget_allocate_segment(td, uap, mode);
 801         } else {
 802                 segnum = shm_find_segment_by_key(td->td_ucred->cr_prison,
 803                     uap->key);
 804                 if (segnum >= 0)
 805                         error = shmget_existing(td, uap, mode, segnum);
 806                 else if ((uap->shmflg & IPC_CREAT) == 0)
 807                         error = ENOENT;
 808                 else
 809                         error = shmget_allocate_segment(td, uap, mode);
 810         }
 811         SYSVSHM_UNLOCK();
 812         return (error);
 813 }
 814
 815 static void
 816 shmfork_myhook(struct proc *p1, struct proc *p2)
 817 {
 818         struct shmmap_state *shmmap_s;
 819         size_t size;
 820         int i;
 821
 822         SYSVSHM_LOCK();
 823         size = shminfo.shmseg * sizeof(struct shmmap_state);
 824         shmmap_s = malloc(size, M_SHM, M_WAITOK);
 825         bcopy(p1->p_vmspace->vm_shm, shmmap_s, size);
 826         p2->p_vmspace->vm_shm = shmmap_s;
 827         for (i = 0; i < shminfo.shmseg; i++, shmmap_s++) {
 828                 if (shmmap_s->shmid != -1) {
 829                         KASSERT(IPCID_TO_IX(shmmap_s->shmid) >= 0 &&
 830                             IPCID_TO_IX(shmmap_s->shmid) < shmalloced,
 831                             ("segnum %d shmalloced %d",
 832                             IPCID_TO_IX(shmmap_s->shmid), shmalloced));
 833                         shmsegs[IPCID_TO_IX(shmmap_s->shmid)].u.shm_nattch++;
 834                 }
 835         }
 836         SYSVSHM_UNLOCK();
 837 }
 838
 839 static void
 840 shmexit_myhook(struct vmspace *vm)
 841 {
 842         struct shmmap_state *base, *shm;
 843         int i;
 844
 845         base = vm->vm_shm;
 846         if (base != NULL) {
 847                 vm->vm_shm = NULL;
 848                 SYSVSHM_LOCK();
 849                 for (i = 0, shm = base; i < shminfo.shmseg; i++, shm++) {
 850                         if (shm->shmid != -1)
 851                                 shm_delete_mapping(vm, shm);
 852                 }
 853                 SYSVSHM_UNLOCK();
 854                 free(base, M_SHM);
 855         }
 856 }
 857
 858 static void
 859 shmrealloc(void)
 860 {
 861         struct shmid_kernel *newsegs;
 862         int i;
 863
 864         SYSVSHM_ASSERT_LOCKED();
 865
 866         if (shmalloced >= shminfo.shmmni)
 867                 return;
 868
 869         newsegs = malloc(shminfo.shmmni * sizeof(*newsegs), M_SHM,
 870             M_WAITOK | M_ZERO);
 871         for (i = 0; i < shmalloced; i++)
 872                 bcopy(&shmsegs[i], &newsegs[i], sizeof(newsegs[0]));
 873         for (; i < shminfo.shmmni; i++) {
 874                 newsegs[i].u.shm_perm.mode = SHMSEG_FREE;
 875                 newsegs[i].u.shm_perm.seq = 0;
 876 #ifdef MAC
 877                 mac_sysvshm_init(&newsegs[i]);
 878 #endif
 879         }
 880         free(shmsegs, M_SHM);
 881         shmsegs = newsegs;
 882         shmalloced = shminfo.shmmni;
 883 }
 884
 885 static struct syscall_helper_data shm_syscalls[] = {
 886         SYSCALL_INIT_HELPER(shmat),
 887         SYSCALL_INIT_HELPER(shmctl),
 888         SYSCALL_INIT_HELPER(shmdt),
 889         SYSCALL_INIT_HELPER(shmget),
 890 #if defined(COMPAT_FREEBSD4) || defined(COMPAT_FREEBSD5) || \
 891     defined(COMPAT_FREEBSD6) || defined(COMPAT_FREEBSD7)
 892         SYSCALL_INIT_HELPER_COMPAT(freebsd7_shmctl),
 893 #endif
 894 #if defined(__i386__) && (defined(COMPAT_FREEBSD4) || defined(COMPAT_43))
 895         SYSCALL_INIT_HELPER(shmsys),
 896 #endif
 897         SYSCALL_INIT_LAST
 898 };
 899
 900 #ifdef COMPAT_FREEBSD32
 901 #include <compat/freebsd32/freebsd32.h>
 902 #include <compat/freebsd32/freebsd32_ipc.h>
 903 #include <compat/freebsd32/freebsd32_proto.h>
 904 #include <compat/freebsd32/freebsd32_signal.h>
 905 #include <compat/freebsd32/freebsd32_syscall.h>
 906 #include <compat/freebsd32/freebsd32_util.h>
 907
 908 static struct syscall_helper_data shm32_syscalls[] = {
 909         SYSCALL32_INIT_HELPER_COMPAT(shmat),
 910         SYSCALL32_INIT_HELPER_COMPAT(shmdt),
 911         SYSCALL32_INIT_HELPER_COMPAT(shmget),
 912         SYSCALL32_INIT_HELPER(freebsd32_shmsys),
 913         SYSCALL32_INIT_HELPER(freebsd32_shmctl),
 914 #if defined(COMPAT_FREEBSD4) || defined(COMPAT_FREEBSD5) || \
 915     defined(COMPAT_FREEBSD6) || defined(COMPAT_FREEBSD7)
 916         SYSCALL32_INIT_HELPER(freebsd7_freebsd32_shmctl),
 917 #endif
 918         SYSCALL_INIT_LAST
 919 };
 920 #endif
 921
 922 static int
 923 shminit(void)
 924 {
 925         struct prison *pr;
 926         void **rsv;
 927         int i, error;
 928         osd_method_t methods[PR_MAXMETHOD] = {
 929             [PR_METHOD_CHECK] =         shm_prison_check,
 930             [PR_METHOD_SET] =           shm_prison_set,
 931             [PR_METHOD_GET] =           shm_prison_get,
 932             [PR_METHOD_REMOVE] =        shm_prison_remove,
 933         };
 934
 935 #ifndef BURN_BRIDGES
 936         if (TUNABLE_ULONG_FETCH("kern.ipc.shmmaxpgs", &shminfo.shmall) != 0)
 937                 printf("kern.ipc.shmmaxpgs is now called kern.ipc.shmall!\n");
 938 #endif
 939         if (shminfo.shmmax == SHMMAX) {
 940                 /* Initialize shmmax dealing with possible overflow. */
 941                 for (i = PAGE_SIZE; i != 0; i--) {
 942                         shminfo.shmmax = shminfo.shmall * i;
 943                         if ((shminfo.shmmax / shminfo.shmall) == (u_long)i)
 944                                 break;
 945                 }
 946         }
 947         shmalloced = shminfo.shmmni;
 948         shmsegs = malloc(shmalloced * sizeof(shmsegs[0]), M_SHM,
 949             M_WAITOK|M_ZERO);
 950         for (i = 0; i < shmalloced; i++) {
 951                 shmsegs[i].u.shm_perm.mode = SHMSEG_FREE;
 952                 shmsegs[i].u.shm_perm.seq = 0;
 953 #ifdef MAC
 954                 mac_sysvshm_init(&shmsegs[i]);
 955 #endif
 956         }
 957         shm_last_free = 0;
 958         shm_nused = 0;
 959         shm_committed = 0;
 960         sx_init(&sysvshmsx, "sysvshmsx");
 961         shmexit_hook = &shmexit_myhook;
 962         shmfork_hook = &shmfork_myhook;
 963
 964         /* Set current prisons according to their allow.sysvipc. */
 965         shm_prison_slot = osd_jail_register(NULL, methods);
 966         rsv = osd_reserve(shm_prison_slot);
 967         prison_lock(&prison0);
 968         (void)osd_jail_set_reserved(&prison0, shm_prison_slot, rsv, &prison0);
 969         prison_unlock(&prison0);
 970         rsv = NULL;
 971         sx_slock(&allprison_lock);
 972         TAILQ_FOREACH(pr, &allprison, pr_list) {
 973                 if (rsv == NULL)
 974                         rsv = osd_reserve(shm_prison_slot);
 975                 prison_lock(pr);
 976                 if ((pr->pr_allow & PR_ALLOW_SYSVIPC) && pr->pr_ref > 0) {
 977                         (void)osd_jail_set_reserved(pr, shm_prison_slot, rsv,
 978                             &prison0);
 979                         rsv = NULL;
 980                 }
 981                 prison_unlock(pr);
 982         }
 983         if (rsv != NULL)
 984                 osd_free_reserved(rsv);
 985         sx_sunlock(&allprison_lock);
 986
 987         error = syscall_helper_register(shm_syscalls, SY_THR_STATIC_KLD);
 988         if (error != 0)
 989                 return (error);
 990 #ifdef COMPAT_FREEBSD32
 991         error = syscall32_helper_register(shm32_syscalls, SY_THR_STATIC_KLD);
 992         if (error != 0)
 993                 return (error);
 994 #endif
 995         return (0);
 996 }
 997
 998 static int
 999 shmunload(void)
1000 {
1001         int i;
1002
1003         if (shm_nused > 0)
1004                 return (EBUSY);
1005
1006 #ifdef COMPAT_FREEBSD32
1007         syscall32_helper_unregister(shm32_syscalls);
1008 #endif
1009         syscall_helper_unregister(shm_syscalls);
1010         if (shm_prison_slot != 0)
1011                 osd_jail_deregister(shm_prison_slot);
1012
1013         for (i = 0; i < shmalloced; i++) {
1014 #ifdef MAC
1015                 mac_sysvshm_destroy(&shmsegs[i]);
1016 #endif
1017                 /*
1018                  * Objects might be still mapped into the processes
1019                  * address spaces.  Actual free would happen on the
1020                  * last mapping destruction.
1021                  */
1022                 if (shmsegs[i].u.shm_perm.mode != SHMSEG_FREE)
1023                         vm_object_deallocate(shmsegs[i].object);
1024         }
1025         free(shmsegs, M_SHM);
1026         shmexit_hook = NULL;
1027         shmfork_hook = NULL;
1028         sx_destroy(&sysvshmsx);
1029         return (0);
1030 }
1031
1032 static int
1033 sysctl_shmsegs(SYSCTL_HANDLER_ARGS)
1034 {
1035         struct shmid_kernel tshmseg;
1036 #ifdef COMPAT_FREEBSD32
1037         struct shmid_kernel32 tshmseg32;
1038 #endif
1039         struct prison *pr, *rpr;
1040         void *outaddr;
1041         size_t outsize;
1042         int error, i;
1043
1044         SYSVSHM_LOCK();
1045         pr = req->td->td_ucred->cr_prison;
1046         rpr = shm_find_prison(req->td->td_ucred);
1047         error = 0;
1048         for (i = 0; i < shmalloced; i++) {
1049                 if ((shmsegs[i].u.shm_perm.mode & SHMSEG_ALLOCATED) == 0 ||
1050                     rpr == NULL || shm_prison_cansee(rpr, &shmsegs[i]) != 0) {
1051                         bzero(&tshmseg, sizeof(tshmseg));
1052                         tshmseg.u.shm_perm.mode = SHMSEG_FREE;
1053                 } else {
1054                         tshmseg = shmsegs[i];
1055                         if (tshmseg.cred->cr_prison != pr)
1056                                 tshmseg.u.shm_perm.key = IPC_PRIVATE;
1057                 }
1058 #ifdef COMPAT_FREEBSD32
1059                 if (SV_CURPROC_FLAG(SV_ILP32)) {
1060                         bzero(&tshmseg32, sizeof(tshmseg32));
1061                         freebsd32_ipcperm_out(&tshmseg.u.shm_perm,
1062                             &tshmseg32.u.shm_perm);
1063                         CP(tshmseg, tshmseg32, u.shm_segsz);
1064                         CP(tshmseg, tshmseg32, u.shm_lpid);
1065                         CP(tshmseg, tshmseg32, u.shm_cpid);
1066                         CP(tshmseg, tshmseg32, u.shm_nattch);
1067                         CP(tshmseg, tshmseg32, u.shm_atime);
1068                         CP(tshmseg, tshmseg32, u.shm_dtime);
1069                         CP(tshmseg, tshmseg32, u.shm_ctime);
1070                         /* Don't copy object, label, or cred */
1071                         outaddr = &tshmseg32;
1072                         outsize = sizeof(tshmseg32);
1073                 } else
1074 #endif
1075                 {
1076                         tshmseg.object = NULL;
1077                         tshmseg.label = NULL;
1078                         tshmseg.cred = NULL;
1079                         outaddr = &tshmseg;
1080                         outsize = sizeof(tshmseg);
1081                 }
1082                 error = SYSCTL_OUT(req, outaddr, outsize);
1083                 if (error != 0)
1084                         break;
1085         }
1086         SYSVSHM_UNLOCK();
1087         return (error);
1088 }
1089
1090 static int
1091 shm_prison_check(void *obj, void *data)
1092 {
1093         struct prison *pr = obj;
1094         struct prison *prpr;
1095         struct vfsoptlist *opts = data;
1096         int error, jsys;
1097
1098         /*
1099          * sysvshm is a jailsys integer.
1100          * It must be "disable" if the parent jail is disabled.
1101          */
1102         error = vfs_copyopt(opts, "sysvshm", &jsys, sizeof(jsys));
1103         if (error != ENOENT) {
1104                 if (error != 0)
1105                         return (error);
1106                 switch (jsys) {
1107                 case JAIL_SYS_DISABLE:
1108                         break;
1109                 case JAIL_SYS_NEW:
1110                 case JAIL_SYS_INHERIT:
1111                         prison_lock(pr->pr_parent);
1112                         prpr = osd_jail_get(pr->pr_parent, shm_prison_slot);
1113                         prison_unlock(pr->pr_parent);
1114                         if (prpr == NULL)
1115                                 return (EPERM);
1116                         break;
1117                 default:
1118                         return (EINVAL);
1119                 }
1120         }
1121
1122         return (0);
1123 }
1124
1125 static int
1126 shm_prison_set(void *obj, void *data)
1127 {
1128         struct prison *pr = obj;
1129         struct prison *tpr, *orpr, *nrpr, *trpr;
1130         struct vfsoptlist *opts = data;
1131         void *rsv;
1132         int jsys, descend;
1133
1134         /*
1135          * sysvshm controls which jail is the root of the associated segments
1136          * (this jail or same as the parent), or if the feature is available
1137          * at all.
1138          */
1139         if (vfs_copyopt(opts, "sysvshm", &jsys, sizeof(jsys)) == ENOENT)
1140                 jsys = vfs_flagopt(opts, "allow.sysvipc", NULL, 0)
1141                     ? JAIL_SYS_INHERIT
1142                     : vfs_flagopt(opts, "allow.nosysvipc", NULL, 0)
1143                     ? JAIL_SYS_DISABLE
1144                     : -1;
1145         if (jsys == JAIL_SYS_DISABLE) {
1146                 prison_lock(pr);
1147                 orpr = osd_jail_get(pr, shm_prison_slot);
1148                 if (orpr != NULL)
1149                         osd_jail_del(pr, shm_prison_slot);
1150                 prison_unlock(pr);
1151                 if (orpr != NULL) {
1152                         if (orpr == pr)
1153                                 shm_prison_cleanup(pr);
1154                         /* Disable all child jails as well. */
1155                         FOREACH_PRISON_DESCENDANT(pr, tpr, descend) {
1156                                 prison_lock(tpr);
1157                                 trpr = osd_jail_get(tpr, shm_prison_slot);
1158                                 if (trpr != NULL) {
1159                                         osd_jail_del(tpr, shm_prison_slot);
1160                                         prison_unlock(tpr);
1161                                         if (trpr == tpr)
1162                                                 shm_prison_cleanup(tpr);
1163                                 } else {
1164                                         prison_unlock(tpr);
1165                                         descend = 0;
1166                                 }
1167                         }
1168                 }
1169         } else if (jsys != -1) {
1170                 if (jsys == JAIL_SYS_NEW)
1171                         nrpr = pr;
1172                 else {
1173                         prison_lock(pr->pr_parent);
1174                         nrpr = osd_jail_get(pr->pr_parent, shm_prison_slot);
1175                         prison_unlock(pr->pr_parent);
1176                 }
1177                 rsv = osd_reserve(shm_prison_slot);
1178                 prison_lock(pr);
1179                 orpr = osd_jail_get(pr, shm_prison_slot);
1180                 if (orpr != nrpr)
1181                         (void)osd_jail_set_reserved(pr, shm_prison_slot, rsv,
1182                             nrpr);
1183                 else
1184                         osd_free_reserved(rsv);
1185                 prison_unlock(pr);
1186                 if (orpr != nrpr) {
1187                         if (orpr == pr)
1188                                 shm_prison_cleanup(pr);
1189                         if (orpr != NULL) {
1190                                 /* Change child jails matching the old root, */
1191                                 FOREACH_PRISON_DESCENDANT(pr, tpr, descend) {
1192                                         prison_lock(tpr);
1193                                         trpr = osd_jail_get(tpr,
1194                                             shm_prison_slot);
1195                                         if (trpr == orpr) {
1196                                                 (void)osd_jail_set(tpr,
1197                                                     shm_prison_slot, nrpr);
1198                                                 prison_unlock(tpr);
1199                                                 if (trpr == tpr)
1200                                                         shm_prison_cleanup(tpr);
1201                                         } else {
1202                                                 prison_unlock(tpr);
1203                                                 descend = 0;
1204                                         }
1205                                 }
1206                         }
1207                 }
1208         }
1209
1210         return (0);
1211 }
1212
1213 static int
1214 shm_prison_get(void *obj, void *data)
1215 {
1216         struct prison *pr = obj;
1217         struct prison *rpr;
1218         struct vfsoptlist *opts = data;
1219         int error, jsys;
1220
1221         /* Set sysvshm based on the jail's root prison. */
1222         prison_lock(pr);
1223         rpr = osd_jail_get(pr, shm_prison_slot);
1224         prison_unlock(pr);
1225         jsys = rpr == NULL ? JAIL_SYS_DISABLE
1226             : rpr == pr ? JAIL_SYS_NEW : JAIL_SYS_INHERIT;
1227         error = vfs_setopt(opts, "sysvshm", &jsys, sizeof(jsys));
1228         if (error == ENOENT)
1229                 error = 0;
1230         return (error);
1231 }
1232
1233 static int
1234 shm_prison_remove(void *obj, void *data __unused)
1235 {
1236         struct prison *pr = obj;
1237         struct prison *rpr;
1238
1239         SYSVSHM_LOCK();
1240         prison_lock(pr);
1241         rpr = osd_jail_get(pr, shm_prison_slot);
1242         prison_unlock(pr);
1243         if (rpr == pr)
1244                 shm_prison_cleanup(pr);
1245         SYSVSHM_UNLOCK();
1246         return (0);
1247 }
1248
1249 static void
1250 shm_prison_cleanup(struct prison *pr)
1251 {
1252         struct shmid_kernel *shmseg;
1253         int i;
1254
1255         /* Remove any segments that belong to this jail. */
1256         for (i = 0; i < shmalloced; i++) {
1257                 shmseg = &shmsegs[i];
1258                 if ((shmseg->u.shm_perm.mode & SHMSEG_ALLOCATED) &&
1259                     shmseg->cred != NULL && shmseg->cred->cr_prison == pr) {
1260                         shm_remove(shmseg, i);
1261                 }
1262         }
1263 }
1264
1265 SYSCTL_JAIL_PARAM_SYS_NODE(sysvshm, CTLFLAG_RW, "SYSV shared memory");
1266
1267 #if defined(__i386__) && (defined(COMPAT_FREEBSD4) || defined(COMPAT_43))
1268 struct oshmid_ds {
1269         struct  ipc_perm_old shm_perm;  /* operation perms */
1270         int     shm_segsz;              /* size of segment (bytes) */
1271         u_short shm_cpid;               /* pid, creator */
1272         u_short shm_lpid;               /* pid, last operation */
1273         short   shm_nattch;             /* no. of current attaches */
1274         time_t  shm_atime;              /* last attach time */
1275         time_t  shm_dtime;              /* last detach time */
1276         time_t  shm_ctime;              /* last change time */
1277         void    *shm_handle;            /* internal handle for shm segment */
1278 };
1279
1280 struct oshmctl_args {
1281         int shmid;
1282         int cmd;
1283         struct oshmid_ds *ubuf;
1284 };
1285
1286 static int
1287 oshmctl(struct thread *td, struct oshmctl_args *uap)
1288 {
1289 #ifdef COMPAT_43
1290         int error = 0;
1291         struct prison *rpr;
1292         struct shmid_kernel *shmseg;
1293         struct oshmid_ds outbuf;
1294
1295         rpr = shm_find_prison(td->td_ucred);
1296         if (rpr == NULL)
1297                 return (ENOSYS);
1298         if (uap->cmd != IPC_STAT) {
1299                 return (freebsd7_shmctl(td,
1300                     (struct freebsd7_shmctl_args *)uap));
1301         }
1302         SYSVSHM_LOCK();
1303         shmseg = shm_find_segment(rpr, uap->shmid, true);
1304         if (shmseg == NULL) {
1305                 SYSVSHM_UNLOCK();
1306                 return (EINVAL);
1307         }
1308         error = ipcperm(td, &shmseg->u.shm_perm, IPC_R);
1309         if (error != 0) {
1310                 SYSVSHM_UNLOCK();
1311                 return (error);
1312         }
1313 #ifdef MAC
1314         error = mac_sysvshm_check_shmctl(td->td_ucred, shmseg, uap->cmd);
1315         if (error != 0) {
1316                 SYSVSHM_UNLOCK();
1317                 return (error);
1318         }
1319 #endif
1320         ipcperm_new2old(&shmseg->u.shm_perm, &outbuf.shm_perm);
1321         outbuf.shm_segsz = shmseg->u.shm_segsz;
1322         outbuf.shm_cpid = shmseg->u.shm_cpid;
1323         outbuf.shm_lpid = shmseg->u.shm_lpid;
1324         outbuf.shm_nattch = shmseg->u.shm_nattch;
1325         outbuf.shm_atime = shmseg->u.shm_atime;
1326         outbuf.shm_dtime = shmseg->u.shm_dtime;
1327         outbuf.shm_ctime = shmseg->u.shm_ctime;
1328         outbuf.shm_handle = shmseg->object;
1329         SYSVSHM_UNLOCK();
1330         return (copyout(&outbuf, uap->ubuf, sizeof(outbuf)));
1331 #else
1332         return (EINVAL);
1333 #endif
1334 }
1335
1336 /* XXX casting to (sy_call_t *) is bogus, as usual. */
1337 static sy_call_t *shmcalls[] = {
1338         (sy_call_t *)sys_shmat, (sy_call_t *)oshmctl,
1339         (sy_call_t *)sys_shmdt, (sy_call_t *)sys_shmget,
1340         (sy_call_t *)freebsd7_shmctl
1341 };
1342
1343 #ifndef _SYS_SYSPROTO_H_
1344 /* XXX actually varargs. */
1345 struct shmsys_args {
1346         int     which;
1347         int     a2;
1348         int     a3;
1349         int     a4;
1350 };
1351 #endif
1352 int
1353 sys_shmsys(struct thread *td, struct shmsys_args *uap)
1354 {
1355
1356         AUDIT_ARG_SVIPC_WHICH(uap->which);
1357         if (uap->which < 0 || uap->which >= nitems(shmcalls))
1358                 return (EINVAL);
1359         return ((*shmcalls[uap->which])(td, &uap->a2));
1360 }
1361
1362 #endif  /* i386 && (COMPAT_FREEBSD4 || COMPAT_43) */
1363
1364 #ifdef COMPAT_FREEBSD32
1365
1366 int
1367 freebsd32_shmsys(struct thread *td, struct freebsd32_shmsys_args *uap)
1368 {
1369
1370 #if defined(COMPAT_FREEBSD4) || defined(COMPAT_FREEBSD5) || \
1371     defined(COMPAT_FREEBSD6) || defined(COMPAT_FREEBSD7)
1372         AUDIT_ARG_SVIPC_WHICH(uap->which);
1373         switch (uap->which) {
1374         case 0: {       /* shmat */
1375                 struct shmat_args ap;
1376
1377                 ap.shmid = uap->a2;
1378                 ap.shmaddr = PTRIN(uap->a3);
1379                 ap.shmflg = uap->a4;
1380                 return (sysent[SYS_shmat].sy_call(td, &ap));
1381         }
1382         case 2: {       /* shmdt */
1383                 struct shmdt_args ap;
1384
1385                 ap.shmaddr = PTRIN(uap->a2);
1386                 return (sysent[SYS_shmdt].sy_call(td, &ap));
1387         }
1388         case 3: {       /* shmget */
1389                 struct shmget_args ap;
1390
1391                 ap.key = uap->a2;
1392                 ap.size = uap->a3;
1393                 ap.shmflg = uap->a4;
1394                 return (sysent[SYS_shmget].sy_call(td, &ap));
1395         }
1396         case 4: {       /* shmctl */
1397                 struct freebsd7_freebsd32_shmctl_args ap;
1398
1399                 ap.shmid = uap->a2;
1400                 ap.cmd = uap->a3;
1401                 ap.buf = PTRIN(uap->a4);
1402                 return (freebsd7_freebsd32_shmctl(td, &ap));
1403         }
1404         case 1:         /* oshmctl */
1405         default:
1406                 return (EINVAL);
1407         }
1408 #else
1409         return (nosys(td, NULL));
1410 #endif
1411 }
1412
1413 #if defined(COMPAT_FREEBSD4) || defined(COMPAT_FREEBSD5) || \
1414     defined(COMPAT_FREEBSD6) || defined(COMPAT_FREEBSD7)
1415 int
1416 freebsd7_freebsd32_shmctl(struct thread *td,
1417     struct freebsd7_freebsd32_shmctl_args *uap)
1418 {
1419         int error;
1420         union {
1421                 struct shmid_ds shmid_ds;
1422                 struct shm_info shm_info;
1423                 struct shminfo shminfo;
1424         } u;
1425         union {
1426                 struct shmid_ds32_old shmid_ds32;
1427                 struct shm_info32 shm_info32;
1428                 struct shminfo32 shminfo32;
1429         } u32;
1430         size_t sz;
1431
1432         if (uap->cmd == IPC_SET) {
1433                 if ((error = copyin(uap->buf, &u32.shmid_ds32,
1434                     sizeof(u32.shmid_ds32))))
1435                         goto done;
1436                 freebsd32_ipcperm_old_in(&u32.shmid_ds32.shm_perm,
1437                     &u.shmid_ds.shm_perm);
1438                 CP(u32.shmid_ds32, u.shmid_ds, shm_segsz);
1439                 CP(u32.shmid_ds32, u.shmid_ds, shm_lpid);
1440                 CP(u32.shmid_ds32, u.shmid_ds, shm_cpid);
1441                 CP(u32.shmid_ds32, u.shmid_ds, shm_nattch);
1442                 CP(u32.shmid_ds32, u.shmid_ds, shm_atime);
1443                 CP(u32.shmid_ds32, u.shmid_ds, shm_dtime);
1444                 CP(u32.shmid_ds32, u.shmid_ds, shm_ctime);
1445         }
1446
1447         error = kern_shmctl(td, uap->shmid, uap->cmd, (void *)&u, &sz);
1448         if (error)
1449                 goto done;
1450
1451         /* Cases in which we need to copyout */
1452         switch (uap->cmd) {
1453         case IPC_INFO:
1454                 CP(u.shminfo, u32.shminfo32, shmmax);
1455                 CP(u.shminfo, u32.shminfo32, shmmin);
1456                 CP(u.shminfo, u32.shminfo32, shmmni);
1457                 CP(u.shminfo, u32.shminfo32, shmseg);
1458                 CP(u.shminfo, u32.shminfo32, shmall);
1459                 error = copyout(&u32.shminfo32, uap->buf,
1460                     sizeof(u32.shminfo32));
1461                 break;
1462         case SHM_INFO:
1463                 CP(u.shm_info, u32.shm_info32, used_ids);
1464                 CP(u.shm_info, u32.shm_info32, shm_rss);
1465                 CP(u.shm_info, u32.shm_info32, shm_tot);
1466                 CP(u.shm_info, u32.shm_info32, shm_swp);
1467                 CP(u.shm_info, u32.shm_info32, swap_attempts);
1468                 CP(u.shm_info, u32.shm_info32, swap_successes);
1469                 error = copyout(&u32.shm_info32, uap->buf,
1470                     sizeof(u32.shm_info32));
1471                 break;
1472         case SHM_STAT:
1473         case IPC_STAT:
1474                 freebsd32_ipcperm_old_out(&u.shmid_ds.shm_perm,
1475                     &u32.shmid_ds32.shm_perm);
1476                 if (u.shmid_ds.shm_segsz > INT32_MAX)
1477                         u32.shmid_ds32.shm_segsz = INT32_MAX;
1478                 else
1479                         CP(u.shmid_ds, u32.shmid_ds32, shm_segsz);
1480                 CP(u.shmid_ds, u32.shmid_ds32, shm_lpid);
1481                 CP(u.shmid_ds, u32.shmid_ds32, shm_cpid);
1482                 CP(u.shmid_ds, u32.shmid_ds32, shm_nattch);
1483                 CP(u.shmid_ds, u32.shmid_ds32, shm_atime);
1484                 CP(u.shmid_ds, u32.shmid_ds32, shm_dtime);
1485                 CP(u.shmid_ds, u32.shmid_ds32, shm_ctime);
1486                 u32.shmid_ds32.shm_internal = 0;
1487                 error = copyout(&u32.shmid_ds32, uap->buf,
1488                     sizeof(u32.shmid_ds32));
1489                 break;
1490         }
1491
1492 done:
1493         if (error) {
1494                 /* Invalidate the return value */
1495                 td->td_retval[0] = -1;
1496         }
1497         return (error);
1498 }
1499 #endif
1500
1501 int
1502 freebsd32_shmctl(struct thread *td, struct freebsd32_shmctl_args *uap)
1503 {
1504         int error;
1505         union {
1506                 struct shmid_ds shmid_ds;
1507                 struct shm_info shm_info;
1508                 struct shminfo shminfo;
1509         } u;
1510         union {
1511                 struct shmid_ds32 shmid_ds32;
1512                 struct shm_info32 shm_info32;
1513                 struct shminfo32 shminfo32;
1514         } u32;
1515         size_t sz;
1516
1517         if (uap->cmd == IPC_SET) {
1518                 if ((error = copyin(uap->buf, &u32.shmid_ds32,
1519                     sizeof(u32.shmid_ds32))))
1520                         goto done;
1521                 freebsd32_ipcperm_in(&u32.shmid_ds32.shm_perm,
1522                     &u.shmid_ds.shm_perm);
1523                 CP(u32.shmid_ds32, u.shmid_ds, shm_segsz);
1524                 CP(u32.shmid_ds32, u.shmid_ds, shm_lpid);
1525                 CP(u32.shmid_ds32, u.shmid_ds, shm_cpid);
1526                 CP(u32.shmid_ds32, u.shmid_ds, shm_nattch);
1527                 CP(u32.shmid_ds32, u.shmid_ds, shm_atime);
1528                 CP(u32.shmid_ds32, u.shmid_ds, shm_dtime);
1529                 CP(u32.shmid_ds32, u.shmid_ds, shm_ctime);
1530         }
1531
1532         error = kern_shmctl(td, uap->shmid, uap->cmd, (void *)&u, &sz);
1533         if (error)
1534                 goto done;
1535
1536         /* Cases in which we need to copyout */
1537         switch (uap->cmd) {
1538         case IPC_INFO:
1539                 CP(u.shminfo, u32.shminfo32, shmmax);
1540                 CP(u.shminfo, u32.shminfo32, shmmin);
1541                 CP(u.shminfo, u32.shminfo32, shmmni);
1542                 CP(u.shminfo, u32.shminfo32, shmseg);
1543                 CP(u.shminfo, u32.shminfo32, shmall);
1544                 error = copyout(&u32.shminfo32, uap->buf,
1545                     sizeof(u32.shminfo32));
1546                 break;
1547         case SHM_INFO:
1548                 CP(u.shm_info, u32.shm_info32, used_ids);
1549                 CP(u.shm_info, u32.shm_info32, shm_rss);
1550                 CP(u.shm_info, u32.shm_info32, shm_tot);
1551                 CP(u.shm_info, u32.shm_info32, shm_swp);
1552                 CP(u.shm_info, u32.shm_info32, swap_attempts);
1553                 CP(u.shm_info, u32.shm_info32, swap_successes);
1554                 error = copyout(&u32.shm_info32, uap->buf,
1555                     sizeof(u32.shm_info32));
1556                 break;
1557         case SHM_STAT:
1558         case IPC_STAT:
1559                 freebsd32_ipcperm_out(&u.shmid_ds.shm_perm,
1560                     &u32.shmid_ds32.shm_perm);
1561                 if (u.shmid_ds.shm_segsz > INT32_MAX)
1562                         u32.shmid_ds32.shm_segsz = INT32_MAX;
1563                 else
1564                         CP(u.shmid_ds, u32.shmid_ds32, shm_segsz);
1565                 CP(u.shmid_ds, u32.shmid_ds32, shm_lpid);
1566                 CP(u.shmid_ds, u32.shmid_ds32, shm_cpid);
1567                 CP(u.shmid_ds, u32.shmid_ds32, shm_nattch);
1568                 CP(u.shmid_ds, u32.shmid_ds32, shm_atime);
1569                 CP(u.shmid_ds, u32.shmid_ds32, shm_dtime);
1570                 CP(u.shmid_ds, u32.shmid_ds32, shm_ctime);
1571                 error = copyout(&u32.shmid_ds32, uap->buf,
1572                     sizeof(u32.shmid_ds32));
1573                 break;
1574         }
1575
1576 done:
1577         if (error) {
1578                 /* Invalidate the return value */
1579                 td->td_retval[0] = -1;
1580         }
1581         return (error);
1582 }
1583 #endif
1584
1585 #if defined(COMPAT_FREEBSD4) || defined(COMPAT_FREEBSD5) || \
1586     defined(COMPAT_FREEBSD6) || defined(COMPAT_FREEBSD7)
1587
1588 #ifndef CP
1589 #define CP(src, dst, fld)       do { (dst).fld = (src).fld; } while (0)
1590 #endif
1591
1592 #ifndef _SYS_SYSPROTO_H_
1593 struct freebsd7_shmctl_args {
1594         int shmid;
1595         int cmd;
1596         struct shmid_ds_old *buf;
1597 };
1598 #endif
1599 int
1600 freebsd7_shmctl(struct thread *td, struct freebsd7_shmctl_args *uap)
1601 {
1602         int error;
1603         struct shmid_ds_old old;
1604         struct shmid_ds buf;
1605         size_t bufsz;
1606
1607         /*
1608          * The only reason IPC_INFO, SHM_INFO, SHM_STAT exists is to support
1609          * Linux binaries.  If we see the call come through the FreeBSD ABI,
1610          * return an error back to the user since we do not to support this.
1611          */
1612         if (uap->cmd == IPC_INFO || uap->cmd == SHM_INFO ||
1613             uap->cmd == SHM_STAT)
1614                 return (EINVAL);
1615
1616         /* IPC_SET needs to copyin the buffer before calling kern_shmctl */
1617         if (uap->cmd == IPC_SET) {
1618                 if ((error = copyin(uap->buf, &old, sizeof(old))))
1619                         goto done;
1620                 ipcperm_old2new(&old.shm_perm, &buf.shm_perm);
1621                 CP(old, buf, shm_segsz);
1622                 CP(old, buf, shm_lpid);
1623                 CP(old, buf, shm_cpid);
1624                 CP(old, buf, shm_nattch);
1625                 CP(old, buf, shm_atime);
1626                 CP(old, buf, shm_dtime);
1627                 CP(old, buf, shm_ctime);
1628         }
1629
1630         error = kern_shmctl(td, uap->shmid, uap->cmd, (void *)&buf, &bufsz);
1631         if (error)
1632                 goto done;
1633
1634         /* Cases in which we need to copyout */
1635         switch (uap->cmd) {
1636         case IPC_STAT:
1637                 ipcperm_new2old(&buf.shm_perm, &old.shm_perm);
1638                 if (buf.shm_segsz > INT_MAX)
1639                         old.shm_segsz = INT_MAX;
1640                 else
1641                         CP(buf, old, shm_segsz);
1642                 CP(buf, old, shm_lpid);
1643                 CP(buf, old, shm_cpid);
1644                 if (buf.shm_nattch > SHRT_MAX)
1645                         old.shm_nattch = SHRT_MAX;
1646                 else
1647                         CP(buf, old, shm_nattch);
1648                 CP(buf, old, shm_atime);
1649                 CP(buf, old, shm_dtime);
1650                 CP(buf, old, shm_ctime);
1651                 old.shm_internal = NULL;
1652                 error = copyout(&old, uap->buf, sizeof(old));
1653                 break;
1654         }
1655
1656 done:
1657         if (error) {
1658                 /* Invalidate the return value */
1659                 td->td_retval[0] = -1;
1660         }
1661         return (error);
1662 }
1663
1664 #endif  /* COMPAT_FREEBSD4 || COMPAT_FREEBSD5 || COMPAT_FREEBSD6 ||
1665            COMPAT_FREEBSD7 */
1666
1667 static int
1668 sysvshm_modload(struct module *module, int cmd, void *arg)
1669 {
1670         int error = 0;
1671
1672         switch (cmd) {
1673         case MOD_LOAD:
1674                 error = shminit();
1675                 if (error != 0)
1676                         shmunload();
1677                 break;
1678         case MOD_UNLOAD:
1679                 error = shmunload();
1680                 break;
1681         case MOD_SHUTDOWN:
1682                 break;
1683         default:
1684                 error = EINVAL;
1685                 break;
1686         }
1687         return (error);
1688 }
1689
1690 static moduledata_t sysvshm_mod = {
1691         "sysvshm",
1692         &sysvshm_modload,
1693         NULL
1694 };
1695
1696 DECLARE_MODULE(sysvshm, sysvshm_mod, SI_SUB_SYSV_SHM, SI_ORDER_FIRST);
1697 MODULE_VERSION(sysvshm, 1);