1 /*- 2 * SPDX-License-Identifier: BSD-4-Clause AND BSD-2-Clause 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_sysvipc.h" 75 76 #include <sys/param.h> 77 #include <sys/systm.h> 78 #include <sys/abi_compat.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 #define SHMSEG_FREE 0x0200 115 #define SHMSEG_REMOVED 0x0400 116 #define SHMSEG_ALLOCATED 0x0800 117 118 static int shm_last_free, shm_nused, shmalloced; 119 vm_size_t shm_committed; 120 static struct shmid_kernel *shmsegs; 121 static unsigned shm_prison_slot; 122 123 struct shmmap_state { 124 vm_offset_t va; 125 int shmid; 126 }; 127 128 static void shm_deallocate_segment(struct shmid_kernel *); 129 static int shm_find_segment_by_key(struct prison *, key_t); 130 static struct shmid_kernel *shm_find_segment(struct prison *, int, bool); 131 static int shm_delete_mapping(struct vmspace *vm, struct shmmap_state *); 132 static int shmget_allocate_segment(struct thread *td, key_t key, size_t size, 133 int mode); 134 static int shmget_existing(struct thread *td, size_t size, int shmflg, 135 int mode, int segnum); 136 static void shmrealloc(void); 137 static int shminit(void); 138 static int sysvshm_modload(struct module *, int, void *); 139 static int shmunload(void); 140 #ifndef SYSVSHM 141 static void shmexit_myhook(struct vmspace *vm); 142 static void shmfork_myhook(struct proc *p1, struct proc *p2); 143 #endif 144 static int sysctl_shmsegs(SYSCTL_HANDLER_ARGS); 145 static void shm_remove(struct shmid_kernel *, int); 146 static struct prison *shm_find_prison(struct ucred *); 147 static int shm_prison_cansee(struct prison *, struct shmid_kernel *); 148 static int shm_prison_check(void *, void *); 149 static int shm_prison_set(void *, void *); 150 static int shm_prison_get(void *, void *); 151 static int shm_prison_remove(void *, void *); 152 static void shm_prison_cleanup(struct prison *); 153 154 /* 155 * Tuneable values. 156 */ 157 #ifndef SHMMAXPGS 158 #define SHMMAXPGS 131072ul /* Note: sysv shared memory is swap backed. */ 159 #endif 160 #ifndef SHMMAX 161 #define SHMMAX (SHMMAXPGS*PAGE_SIZE) 162 #endif 163 #ifndef SHMMIN 164 #define SHMMIN 1 165 #endif 166 #ifndef SHMMNI 167 #define SHMMNI 192 168 #endif 169 #ifndef SHMSEG 170 #define SHMSEG 128 171 #endif 172 #ifndef SHMALL 173 #define SHMALL (SHMMAXPGS) 174 #endif 175 176 struct shminfo shminfo = { 177 .shmmax = SHMMAX, 178 .shmmin = SHMMIN, 179 .shmmni = SHMMNI, 180 .shmseg = SHMSEG, 181 .shmall = SHMALL 182 }; 183 184 static int shm_use_phys; 185 static int shm_allow_removed = 1; 186 187 SYSCTL_ULONG(_kern_ipc, OID_AUTO, shmmax, CTLFLAG_RWTUN, &shminfo.shmmax, 0, 188 "Maximum shared memory segment size"); 189 SYSCTL_ULONG(_kern_ipc, OID_AUTO, shmmin, CTLFLAG_RWTUN, &shminfo.shmmin, 0, 190 "Minimum shared memory segment size"); 191 SYSCTL_ULONG(_kern_ipc, OID_AUTO, shmmni, CTLFLAG_RDTUN, &shminfo.shmmni, 0, 192 "Number of shared memory identifiers"); 193 SYSCTL_ULONG(_kern_ipc, OID_AUTO, shmseg, CTLFLAG_RDTUN, &shminfo.shmseg, 0, 194 "Number of segments per process"); 195 SYSCTL_ULONG(_kern_ipc, OID_AUTO, shmall, CTLFLAG_RWTUN, &shminfo.shmall, 0, 196 "Maximum number of pages available for shared memory"); 197 SYSCTL_INT(_kern_ipc, OID_AUTO, shm_use_phys, CTLFLAG_RWTUN, 198 &shm_use_phys, 0, "Enable/Disable locking of shared memory pages in core"); 199 SYSCTL_INT(_kern_ipc, OID_AUTO, shm_allow_removed, CTLFLAG_RWTUN, 200 &shm_allow_removed, 0, 201 "Enable/Disable attachment to attached segments marked for removal"); 202 SYSCTL_PROC(_kern_ipc, OID_AUTO, shmsegs, CTLTYPE_OPAQUE | CTLFLAG_RD | 203 CTLFLAG_MPSAFE, NULL, 0, sysctl_shmsegs, "", 204 "Array of struct shmid_kernel for each potential shared memory segment"); 205 206 static struct sx sysvshmsx; 207 #define SYSVSHM_LOCK() sx_xlock(&sysvshmsx) 208 #define SYSVSHM_UNLOCK() sx_xunlock(&sysvshmsx) 209 #define SYSVSHM_ASSERT_LOCKED() sx_assert(&sysvshmsx, SA_XLOCKED) 210 211 static int 212 shm_find_segment_by_key(struct prison *pr, key_t key) 213 { 214 int i; 215 216 for (i = 0; i < shmalloced; i++) 217 if ((shmsegs[i].u.shm_perm.mode & SHMSEG_ALLOCATED) && 218 shmsegs[i].cred != NULL && 219 shmsegs[i].cred->cr_prison == pr && 220 shmsegs[i].u.shm_perm.key == key) 221 return (i); 222 return (-1); 223 } 224 225 /* 226 * Finds segment either by shmid if is_shmid is true, or by segnum if 227 * is_shmid is false. 228 */ 229 static struct shmid_kernel * 230 shm_find_segment(struct prison *rpr, int arg, bool is_shmid) 231 { 232 struct shmid_kernel *shmseg; 233 int segnum; 234 235 segnum = is_shmid ? IPCID_TO_IX(arg) : arg; 236 if (segnum < 0 || segnum >= shmalloced) 237 return (NULL); 238 shmseg = &shmsegs[segnum]; 239 if ((shmseg->u.shm_perm.mode & SHMSEG_ALLOCATED) == 0 || 240 (!shm_allow_removed && 241 (shmseg->u.shm_perm.mode & SHMSEG_REMOVED) != 0) || 242 (is_shmid && shmseg->u.shm_perm.seq != IPCID_TO_SEQ(arg)) || 243 shm_prison_cansee(rpr, shmseg) != 0) 244 return (NULL); 245 return (shmseg); 246 } 247 248 static void 249 shm_deallocate_segment(struct shmid_kernel *shmseg) 250 { 251 vm_size_t size; 252 253 SYSVSHM_ASSERT_LOCKED(); 254 255 vm_object_deallocate(shmseg->object); 256 shmseg->object = NULL; 257 size = round_page(shmseg->u.shm_segsz); 258 shm_committed -= btoc(size); 259 shm_nused--; 260 shmseg->u.shm_perm.mode = SHMSEG_FREE; 261 #ifdef MAC 262 mac_sysvshm_cleanup(shmseg); 263 #endif 264 racct_sub_cred(shmseg->cred, RACCT_NSHM, 1); 265 racct_sub_cred(shmseg->cred, RACCT_SHMSIZE, size); 266 crfree(shmseg->cred); 267 shmseg->cred = NULL; 268 } 269 270 static int 271 shm_delete_mapping(struct vmspace *vm, struct shmmap_state *shmmap_s) 272 { 273 struct shmid_kernel *shmseg; 274 int segnum, result; 275 vm_size_t size; 276 277 SYSVSHM_ASSERT_LOCKED(); 278 segnum = IPCID_TO_IX(shmmap_s->shmid); 279 KASSERT(segnum >= 0 && segnum < shmalloced, 280 ("segnum %d shmalloced %d", segnum, shmalloced)); 281 282 shmseg = &shmsegs[segnum]; 283 size = round_page(shmseg->u.shm_segsz); 284 result = vm_map_remove(&vm->vm_map, shmmap_s->va, shmmap_s->va + size); 285 if (result != KERN_SUCCESS) 286 return (EINVAL); 287 shmmap_s->shmid = -1; 288 shmseg->u.shm_dtime = time_second; 289 if (--shmseg->u.shm_nattch == 0 && 290 (shmseg->u.shm_perm.mode & SHMSEG_REMOVED)) { 291 shm_deallocate_segment(shmseg); 292 shm_last_free = segnum; 293 } 294 return (0); 295 } 296 297 static void 298 shm_remove(struct shmid_kernel *shmseg, int segnum) 299 { 300 301 shmseg->u.shm_perm.key = IPC_PRIVATE; 302 shmseg->u.shm_perm.mode |= SHMSEG_REMOVED; 303 if (shmseg->u.shm_nattch == 0) { 304 shm_deallocate_segment(shmseg); 305 shm_last_free = segnum; 306 } 307 } 308 309 static struct prison * 310 shm_find_prison(struct ucred *cred) 311 { 312 struct prison *pr, *rpr; 313 314 pr = cred->cr_prison; 315 prison_lock(pr); 316 rpr = osd_jail_get(pr, shm_prison_slot); 317 prison_unlock(pr); 318 return rpr; 319 } 320 321 static int 322 shm_prison_cansee(struct prison *rpr, struct shmid_kernel *shmseg) 323 { 324 325 if (shmseg->cred == NULL || 326 !(rpr == shmseg->cred->cr_prison || 327 prison_ischild(rpr, shmseg->cred->cr_prison))) 328 return (EINVAL); 329 return (0); 330 } 331 332 static int 333 kern_shmdt_locked(struct thread *td, const void *shmaddr) 334 { 335 struct proc *p = td->td_proc; 336 struct shmmap_state *shmmap_s; 337 #ifdef MAC 338 int error; 339 #endif 340 int i; 341 342 SYSVSHM_ASSERT_LOCKED(); 343 if (shm_find_prison(td->td_ucred) == NULL) 344 return (ENOSYS); 345 shmmap_s = p->p_vmspace->vm_shm; 346 if (shmmap_s == NULL) 347 return (EINVAL); 348 AUDIT_ARG_SVIPC_ID(shmmap_s->shmid); 349 for (i = 0; i < shminfo.shmseg; i++, shmmap_s++) { 350 if (shmmap_s->shmid != -1 && 351 shmmap_s->va == (vm_offset_t)shmaddr) { 352 break; 353 } 354 } 355 if (i == shminfo.shmseg) 356 return (EINVAL); 357 #ifdef MAC 358 error = mac_sysvshm_check_shmdt(td->td_ucred, 359 &shmsegs[IPCID_TO_IX(shmmap_s->shmid)]); 360 if (error != 0) 361 return (error); 362 #endif 363 return (shm_delete_mapping(p->p_vmspace, shmmap_s)); 364 } 365 366 #ifndef _SYS_SYSPROTO_H_ 367 struct shmdt_args { 368 const void *shmaddr; 369 }; 370 #endif 371 int 372 sys_shmdt(struct thread *td, struct shmdt_args *uap) 373 { 374 int error; 375 376 SYSVSHM_LOCK(); 377 error = kern_shmdt_locked(td, uap->shmaddr); 378 SYSVSHM_UNLOCK(); 379 return (error); 380 } 381 382 static int 383 kern_shmat_locked(struct thread *td, int shmid, const void *shmaddr, 384 int shmflg) 385 { 386 struct prison *rpr; 387 struct proc *p = td->td_proc; 388 struct shmid_kernel *shmseg; 389 struct shmmap_state *shmmap_s; 390 vm_offset_t attach_va; 391 vm_prot_t prot; 392 vm_size_t size; 393 int cow, error, find_space, i, rv; 394 395 AUDIT_ARG_SVIPC_ID(shmid); 396 AUDIT_ARG_VALUE(shmflg); 397 398 SYSVSHM_ASSERT_LOCKED(); 399 rpr = shm_find_prison(td->td_ucred); 400 if (rpr == NULL) 401 return (ENOSYS); 402 shmmap_s = p->p_vmspace->vm_shm; 403 if (shmmap_s == NULL) { 404 shmmap_s = malloc(shminfo.shmseg * sizeof(struct shmmap_state), 405 M_SHM, M_WAITOK); 406 for (i = 0; i < shminfo.shmseg; i++) 407 shmmap_s[i].shmid = -1; 408 KASSERT(p->p_vmspace->vm_shm == NULL, ("raced")); 409 p->p_vmspace->vm_shm = shmmap_s; 410 } 411 shmseg = shm_find_segment(rpr, shmid, true); 412 if (shmseg == NULL) 413 return (EINVAL); 414 error = ipcperm(td, &shmseg->u.shm_perm, 415 (shmflg & SHM_RDONLY) ? IPC_R : IPC_R|IPC_W); 416 if (error != 0) 417 return (error); 418 #ifdef MAC 419 error = mac_sysvshm_check_shmat(td->td_ucred, shmseg, shmflg); 420 if (error != 0) 421 return (error); 422 #endif 423 for (i = 0; i < shminfo.shmseg; i++) { 424 if (shmmap_s->shmid == -1) 425 break; 426 shmmap_s++; 427 } 428 if (i >= shminfo.shmseg) 429 return (EMFILE); 430 size = round_page(shmseg->u.shm_segsz); 431 prot = VM_PROT_READ; 432 cow = MAP_INHERIT_SHARE | MAP_PREFAULT_PARTIAL; 433 if ((shmflg & SHM_RDONLY) == 0) 434 prot |= VM_PROT_WRITE; 435 if (shmaddr != NULL) { 436 if ((shmflg & SHM_RND) != 0) 437 attach_va = rounddown2((vm_offset_t)shmaddr, SHMLBA); 438 else if (((vm_offset_t)shmaddr & (SHMLBA-1)) == 0) 439 attach_va = (vm_offset_t)shmaddr; 440 else 441 return (EINVAL); 442 if ((shmflg & SHM_REMAP) != 0) 443 cow |= MAP_REMAP; 444 find_space = VMFS_NO_SPACE; 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 find_space = VMFS_OPTIMAL_SPACE; 453 } 454 455 vm_object_reference(shmseg->object); 456 rv = vm_map_find(&p->p_vmspace->vm_map, shmseg->object, 0, &attach_va, 457 size, 0, find_space, prot, prot, cow); 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 #ifndef _SYS_SYSPROTO_H_ 613 struct shmctl_args { 614 int shmid; 615 int cmd; 616 struct shmid_ds *buf; 617 }; 618 #endif 619 int 620 sys_shmctl(struct thread *td, struct shmctl_args *uap) 621 { 622 int error; 623 struct shmid_ds buf; 624 size_t bufsz; 625 626 /* 627 * The only reason IPC_INFO, SHM_INFO, SHM_STAT exists is to support 628 * Linux binaries. If we see the call come through the FreeBSD ABI, 629 * return an error back to the user since we do not to support this. 630 */ 631 if (uap->cmd == IPC_INFO || uap->cmd == SHM_INFO || 632 uap->cmd == SHM_STAT) 633 return (EINVAL); 634 635 /* IPC_SET needs to copyin the buffer before calling kern_shmctl */ 636 if (uap->cmd == IPC_SET) { 637 if ((error = copyin(uap->buf, &buf, sizeof(struct shmid_ds)))) 638 goto done; 639 } 640 641 error = kern_shmctl(td, uap->shmid, uap->cmd, (void *)&buf, &bufsz); 642 if (error) 643 goto done; 644 645 /* Cases in which we need to copyout */ 646 switch (uap->cmd) { 647 case IPC_STAT: 648 error = copyout(&buf, uap->buf, bufsz); 649 break; 650 } 651 652 done: 653 if (error) { 654 /* Invalidate the return value */ 655 td->td_retval[0] = -1; 656 } 657 return (error); 658 } 659 660 static int 661 shmget_existing(struct thread *td, size_t size, int shmflg, int mode, 662 int segnum) 663 { 664 struct shmid_kernel *shmseg; 665 #ifdef MAC 666 int error; 667 #endif 668 669 SYSVSHM_ASSERT_LOCKED(); 670 KASSERT(segnum >= 0 && segnum < shmalloced, 671 ("segnum %d shmalloced %d", segnum, shmalloced)); 672 shmseg = &shmsegs[segnum]; 673 if ((shmflg & (IPC_CREAT | IPC_EXCL)) == (IPC_CREAT | IPC_EXCL)) 674 return (EEXIST); 675 #ifdef MAC 676 error = mac_sysvshm_check_shmget(td->td_ucred, shmseg, shmflg); 677 if (error != 0) 678 return (error); 679 #endif 680 if (size != 0 && size > shmseg->u.shm_segsz) 681 return (EINVAL); 682 td->td_retval[0] = IXSEQ_TO_IPCID(segnum, shmseg->u.shm_perm); 683 return (0); 684 } 685 686 static int 687 shmget_allocate_segment(struct thread *td, key_t key, size_t size, int mode) 688 { 689 struct ucred *cred = td->td_ucred; 690 struct shmid_kernel *shmseg; 691 vm_object_t shm_object; 692 int i, segnum; 693 694 SYSVSHM_ASSERT_LOCKED(); 695 696 if (size < shminfo.shmmin || size > shminfo.shmmax) 697 return (EINVAL); 698 if (shm_nused >= shminfo.shmmni) /* Any shmids left? */ 699 return (ENOSPC); 700 size = round_page(size); 701 if (shm_committed + btoc(size) > shminfo.shmall) 702 return (ENOMEM); 703 if (shm_last_free < 0) { 704 shmrealloc(); /* Maybe expand the shmsegs[] array. */ 705 for (i = 0; i < shmalloced; i++) 706 if (shmsegs[i].u.shm_perm.mode & SHMSEG_FREE) 707 break; 708 if (i == shmalloced) 709 return (ENOSPC); 710 segnum = i; 711 } else { 712 segnum = shm_last_free; 713 shm_last_free = -1; 714 } 715 KASSERT(segnum >= 0 && segnum < shmalloced, 716 ("segnum %d shmalloced %d", segnum, shmalloced)); 717 shmseg = &shmsegs[segnum]; 718 #ifdef RACCT 719 if (racct_enable) { 720 PROC_LOCK(td->td_proc); 721 if (racct_add(td->td_proc, RACCT_NSHM, 1)) { 722 PROC_UNLOCK(td->td_proc); 723 return (ENOSPC); 724 } 725 if (racct_add(td->td_proc, RACCT_SHMSIZE, size)) { 726 racct_sub(td->td_proc, RACCT_NSHM, 1); 727 PROC_UNLOCK(td->td_proc); 728 return (ENOMEM); 729 } 730 PROC_UNLOCK(td->td_proc); 731 } 732 #endif 733 734 /* 735 * We make sure that we have allocated a pager before we need 736 * to. 737 */ 738 shm_object = vm_pager_allocate(shm_use_phys ? OBJT_PHYS : OBJT_SWAP, 739 0, size, VM_PROT_DEFAULT, 0, cred); 740 if (shm_object == NULL) { 741 #ifdef RACCT 742 if (racct_enable) { 743 PROC_LOCK(td->td_proc); 744 racct_sub(td->td_proc, RACCT_NSHM, 1); 745 racct_sub(td->td_proc, RACCT_SHMSIZE, size); 746 PROC_UNLOCK(td->td_proc); 747 } 748 #endif 749 return (ENOMEM); 750 } 751 752 shmseg->object = shm_object; 753 shmseg->u.shm_perm.cuid = shmseg->u.shm_perm.uid = cred->cr_uid; 754 shmseg->u.shm_perm.cgid = shmseg->u.shm_perm.gid = cred->cr_gid; 755 shmseg->u.shm_perm.mode = (mode & ACCESSPERMS) | SHMSEG_ALLOCATED; 756 shmseg->u.shm_perm.key = key; 757 shmseg->u.shm_perm.seq = (shmseg->u.shm_perm.seq + 1) & 0x7fff; 758 shmseg->cred = crhold(cred); 759 shmseg->u.shm_segsz = size; 760 shmseg->u.shm_cpid = td->td_proc->p_pid; 761 shmseg->u.shm_lpid = shmseg->u.shm_nattch = 0; 762 shmseg->u.shm_atime = shmseg->u.shm_dtime = 0; 763 #ifdef MAC 764 mac_sysvshm_create(cred, shmseg); 765 #endif 766 shmseg->u.shm_ctime = time_second; 767 shm_committed += btoc(size); 768 shm_nused++; 769 td->td_retval[0] = IXSEQ_TO_IPCID(segnum, shmseg->u.shm_perm); 770 771 return (0); 772 } 773 774 #ifndef _SYS_SYSPROTO_H_ 775 struct shmget_args { 776 key_t key; 777 size_t size; 778 int shmflg; 779 }; 780 #endif 781 int 782 sys_shmget(struct thread *td, struct shmget_args *uap) 783 { 784 int segnum, mode; 785 int error; 786 787 if (shm_find_prison(td->td_ucred) == NULL) 788 return (ENOSYS); 789 mode = uap->shmflg & ACCESSPERMS; 790 SYSVSHM_LOCK(); 791 if (uap->key == IPC_PRIVATE) { 792 error = shmget_allocate_segment(td, uap->key, uap->size, mode); 793 } else { 794 segnum = shm_find_segment_by_key(td->td_ucred->cr_prison, 795 uap->key); 796 if (segnum >= 0) 797 error = shmget_existing(td, uap->size, uap->shmflg, 798 mode, segnum); 799 else if ((uap->shmflg & IPC_CREAT) == 0) 800 error = ENOENT; 801 else 802 error = shmget_allocate_segment(td, uap->key, 803 uap->size, mode); 804 } 805 SYSVSHM_UNLOCK(); 806 return (error); 807 } 808 809 #ifdef SYSVSHM 810 void 811 shmfork(struct proc *p1, struct proc *p2) 812 #else 813 static void 814 shmfork_myhook(struct proc *p1, struct proc *p2) 815 #endif 816 { 817 struct shmmap_state *shmmap_s; 818 size_t size; 819 int i; 820 821 SYSVSHM_LOCK(); 822 size = shminfo.shmseg * sizeof(struct shmmap_state); 823 shmmap_s = malloc(size, M_SHM, M_WAITOK); 824 bcopy(p1->p_vmspace->vm_shm, shmmap_s, size); 825 p2->p_vmspace->vm_shm = shmmap_s; 826 for (i = 0; i < shminfo.shmseg; i++, shmmap_s++) { 827 if (shmmap_s->shmid != -1) { 828 KASSERT(IPCID_TO_IX(shmmap_s->shmid) >= 0 && 829 IPCID_TO_IX(shmmap_s->shmid) < shmalloced, 830 ("segnum %d shmalloced %d", 831 IPCID_TO_IX(shmmap_s->shmid), shmalloced)); 832 shmsegs[IPCID_TO_IX(shmmap_s->shmid)].u.shm_nattch++; 833 } 834 } 835 SYSVSHM_UNLOCK(); 836 } 837 838 #ifdef SYSVSHM 839 void 840 shmexit(struct vmspace *vm) 841 #else 842 static void 843 shmexit_myhook(struct vmspace *vm) 844 #endif 845 { 846 struct shmmap_state *base, *shm; 847 int i; 848 849 base = vm->vm_shm; 850 if (base != NULL) { 851 vm->vm_shm = NULL; 852 SYSVSHM_LOCK(); 853 for (i = 0, shm = base; i < shminfo.shmseg; i++, shm++) { 854 if (shm->shmid != -1) 855 shm_delete_mapping(vm, shm); 856 } 857 SYSVSHM_UNLOCK(); 858 free(base, M_SHM); 859 } 860 } 861 862 static void 863 shmrealloc(void) 864 { 865 struct shmid_kernel *newsegs; 866 int i; 867 868 SYSVSHM_ASSERT_LOCKED(); 869 870 if (shmalloced >= shminfo.shmmni) 871 return; 872 873 newsegs = malloc(shminfo.shmmni * sizeof(*newsegs), M_SHM, 874 M_WAITOK | M_ZERO); 875 for (i = 0; i < shmalloced; i++) 876 bcopy(&shmsegs[i], &newsegs[i], sizeof(newsegs[0])); 877 for (; i < shminfo.shmmni; i++) { 878 newsegs[i].u.shm_perm.mode = SHMSEG_FREE; 879 newsegs[i].u.shm_perm.seq = 0; 880 #ifdef MAC 881 mac_sysvshm_init(&newsegs[i]); 882 #endif 883 } 884 free(shmsegs, M_SHM); 885 shmsegs = newsegs; 886 shmalloced = shminfo.shmmni; 887 } 888 889 static struct syscall_helper_data shm_syscalls[] = { 890 SYSCALL_INIT_HELPER(shmat), 891 SYSCALL_INIT_HELPER(shmctl), 892 SYSCALL_INIT_HELPER(shmdt), 893 SYSCALL_INIT_HELPER(shmget), 894 #if defined(COMPAT_FREEBSD4) || defined(COMPAT_FREEBSD5) || \ 895 defined(COMPAT_FREEBSD6) || defined(COMPAT_FREEBSD7) 896 SYSCALL_INIT_HELPER_COMPAT(freebsd7_shmctl), 897 #endif 898 #if defined(__i386__) && (defined(COMPAT_FREEBSD4) || defined(COMPAT_43)) 899 SYSCALL_INIT_HELPER(shmsys), 900 #endif 901 SYSCALL_INIT_LAST 902 }; 903 904 #ifdef COMPAT_FREEBSD32 905 #include <compat/freebsd32/freebsd32.h> 906 #include <compat/freebsd32/freebsd32_ipc.h> 907 #include <compat/freebsd32/freebsd32_proto.h> 908 #include <compat/freebsd32/freebsd32_signal.h> 909 #include <compat/freebsd32/freebsd32_syscall.h> 910 #include <compat/freebsd32/freebsd32_util.h> 911 912 static struct syscall_helper_data shm32_syscalls[] = { 913 SYSCALL32_INIT_HELPER_COMPAT(shmat), 914 SYSCALL32_INIT_HELPER_COMPAT(shmdt), 915 SYSCALL32_INIT_HELPER_COMPAT(shmget), 916 SYSCALL32_INIT_HELPER(freebsd32_shmsys), 917 SYSCALL32_INIT_HELPER(freebsd32_shmctl), 918 #if defined(COMPAT_FREEBSD4) || defined(COMPAT_FREEBSD5) || \ 919 defined(COMPAT_FREEBSD6) || defined(COMPAT_FREEBSD7) 920 SYSCALL32_INIT_HELPER(freebsd7_freebsd32_shmctl), 921 #endif 922 SYSCALL_INIT_LAST 923 }; 924 #endif 925 926 static int 927 shminit(void) 928 { 929 struct prison *pr; 930 void **rsv; 931 int i, error; 932 osd_method_t methods[PR_MAXMETHOD] = { 933 [PR_METHOD_CHECK] = shm_prison_check, 934 [PR_METHOD_SET] = shm_prison_set, 935 [PR_METHOD_GET] = shm_prison_get, 936 [PR_METHOD_REMOVE] = shm_prison_remove, 937 }; 938 939 #ifndef BURN_BRIDGES 940 if (TUNABLE_ULONG_FETCH("kern.ipc.shmmaxpgs", &shminfo.shmall) != 0) 941 printf("kern.ipc.shmmaxpgs is now called kern.ipc.shmall!\n"); 942 #endif 943 if (shminfo.shmmax == SHMMAX) { 944 /* Initialize shmmax dealing with possible overflow. */ 945 for (i = PAGE_SIZE; i != 0; i--) { 946 shminfo.shmmax = shminfo.shmall * i; 947 if ((shminfo.shmmax / shminfo.shmall) == (u_long)i) 948 break; 949 } 950 } 951 shmalloced = shminfo.shmmni; 952 shmsegs = malloc(shmalloced * sizeof(shmsegs[0]), M_SHM, 953 M_WAITOK|M_ZERO); 954 for (i = 0; i < shmalloced; i++) { 955 shmsegs[i].u.shm_perm.mode = SHMSEG_FREE; 956 shmsegs[i].u.shm_perm.seq = 0; 957 #ifdef MAC 958 mac_sysvshm_init(&shmsegs[i]); 959 #endif 960 } 961 shm_last_free = 0; 962 shm_nused = 0; 963 shm_committed = 0; 964 sx_init(&sysvshmsx, "sysvshmsx"); 965 #ifndef SYSVSHM 966 shmexit_hook = &shmexit_myhook; 967 shmfork_hook = &shmfork_myhook; 968 #endif 969 970 /* Set current prisons according to their allow.sysvipc. */ 971 shm_prison_slot = osd_jail_register(NULL, methods); 972 rsv = osd_reserve(shm_prison_slot); 973 prison_lock(&prison0); 974 (void)osd_jail_set_reserved(&prison0, shm_prison_slot, rsv, &prison0); 975 prison_unlock(&prison0); 976 rsv = NULL; 977 sx_slock(&allprison_lock); 978 TAILQ_FOREACH(pr, &allprison, pr_list) { 979 if (rsv == NULL) 980 rsv = osd_reserve(shm_prison_slot); 981 prison_lock(pr); 982 if (pr->pr_allow & PR_ALLOW_SYSVIPC) { 983 (void)osd_jail_set_reserved(pr, shm_prison_slot, rsv, 984 &prison0); 985 rsv = NULL; 986 } 987 prison_unlock(pr); 988 } 989 if (rsv != NULL) 990 osd_free_reserved(rsv); 991 sx_sunlock(&allprison_lock); 992 993 error = syscall_helper_register(shm_syscalls, SY_THR_STATIC_KLD); 994 if (error != 0) 995 return (error); 996 #ifdef COMPAT_FREEBSD32 997 error = syscall32_helper_register(shm32_syscalls, SY_THR_STATIC_KLD); 998 if (error != 0) 999 return (error); 1000 #endif 1001 return (0); 1002 } 1003 1004 static int 1005 shmunload(void) 1006 { 1007 int i; 1008 1009 if (shm_nused > 0) 1010 return (EBUSY); 1011 1012 #ifdef COMPAT_FREEBSD32 1013 syscall32_helper_unregister(shm32_syscalls); 1014 #endif 1015 syscall_helper_unregister(shm_syscalls); 1016 if (shm_prison_slot != 0) 1017 osd_jail_deregister(shm_prison_slot); 1018 1019 for (i = 0; i < shmalloced; i++) { 1020 #ifdef MAC 1021 mac_sysvshm_destroy(&shmsegs[i]); 1022 #endif 1023 /* 1024 * Objects might be still mapped into the processes 1025 * address spaces. Actual free would happen on the 1026 * last mapping destruction. 1027 */ 1028 if (shmsegs[i].u.shm_perm.mode != SHMSEG_FREE) 1029 vm_object_deallocate(shmsegs[i].object); 1030 } 1031 free(shmsegs, M_SHM); 1032 #ifndef SYSVSHM 1033 shmexit_hook = NULL; 1034 shmfork_hook = NULL; 1035 #endif 1036 sx_destroy(&sysvshmsx); 1037 return (0); 1038 } 1039 1040 static int 1041 sysctl_shmsegs(SYSCTL_HANDLER_ARGS) 1042 { 1043 struct shmid_kernel tshmseg; 1044 #ifdef COMPAT_FREEBSD32 1045 struct shmid_kernel32 tshmseg32; 1046 #endif 1047 struct prison *pr, *rpr; 1048 void *outaddr; 1049 size_t outsize; 1050 int error, i; 1051 1052 SYSVSHM_LOCK(); 1053 pr = req->td->td_ucred->cr_prison; 1054 rpr = shm_find_prison(req->td->td_ucred); 1055 error = 0; 1056 for (i = 0; i < shmalloced; i++) { 1057 if ((shmsegs[i].u.shm_perm.mode & SHMSEG_ALLOCATED) == 0 || 1058 rpr == NULL || shm_prison_cansee(rpr, &shmsegs[i]) != 0) { 1059 bzero(&tshmseg, sizeof(tshmseg)); 1060 tshmseg.u.shm_perm.mode = SHMSEG_FREE; 1061 } else { 1062 tshmseg = shmsegs[i]; 1063 if (tshmseg.cred->cr_prison != pr) 1064 tshmseg.u.shm_perm.key = IPC_PRIVATE; 1065 } 1066 #ifdef COMPAT_FREEBSD32 1067 if (SV_CURPROC_FLAG(SV_ILP32)) { 1068 bzero(&tshmseg32, sizeof(tshmseg32)); 1069 freebsd32_ipcperm_out(&tshmseg.u.shm_perm, 1070 &tshmseg32.u.shm_perm); 1071 CP(tshmseg, tshmseg32, u.shm_segsz); 1072 CP(tshmseg, tshmseg32, u.shm_lpid); 1073 CP(tshmseg, tshmseg32, u.shm_cpid); 1074 CP(tshmseg, tshmseg32, u.shm_nattch); 1075 CP(tshmseg, tshmseg32, u.shm_atime); 1076 CP(tshmseg, tshmseg32, u.shm_dtime); 1077 CP(tshmseg, tshmseg32, u.shm_ctime); 1078 /* Don't copy object, label, or cred */ 1079 outaddr = &tshmseg32; 1080 outsize = sizeof(tshmseg32); 1081 } else 1082 #endif 1083 { 1084 tshmseg.object = NULL; 1085 tshmseg.label = NULL; 1086 tshmseg.cred = NULL; 1087 outaddr = &tshmseg; 1088 outsize = sizeof(tshmseg); 1089 } 1090 error = SYSCTL_OUT(req, outaddr, outsize); 1091 if (error != 0) 1092 break; 1093 } 1094 SYSVSHM_UNLOCK(); 1095 return (error); 1096 } 1097 1098 static int 1099 shm_prison_check(void *obj, void *data) 1100 { 1101 struct prison *pr = obj; 1102 struct prison *prpr; 1103 struct vfsoptlist *opts = data; 1104 int error, jsys; 1105 1106 /* 1107 * sysvshm is a jailsys integer. 1108 * It must be "disable" if the parent jail is disabled. 1109 */ 1110 error = vfs_copyopt(opts, "sysvshm", &jsys, sizeof(jsys)); 1111 if (error != ENOENT) { 1112 if (error != 0) 1113 return (error); 1114 switch (jsys) { 1115 case JAIL_SYS_DISABLE: 1116 break; 1117 case JAIL_SYS_NEW: 1118 case JAIL_SYS_INHERIT: 1119 prison_lock(pr->pr_parent); 1120 prpr = osd_jail_get(pr->pr_parent, shm_prison_slot); 1121 prison_unlock(pr->pr_parent); 1122 if (prpr == NULL) 1123 return (EPERM); 1124 break; 1125 default: 1126 return (EINVAL); 1127 } 1128 } 1129 1130 return (0); 1131 } 1132 1133 static int 1134 shm_prison_set(void *obj, void *data) 1135 { 1136 struct prison *pr = obj; 1137 struct prison *tpr, *orpr, *nrpr, *trpr; 1138 struct vfsoptlist *opts = data; 1139 void *rsv; 1140 int jsys, descend; 1141 1142 /* 1143 * sysvshm controls which jail is the root of the associated segments 1144 * (this jail or same as the parent), or if the feature is available 1145 * at all. 1146 */ 1147 if (vfs_copyopt(opts, "sysvshm", &jsys, sizeof(jsys)) == ENOENT) 1148 jsys = vfs_flagopt(opts, "allow.sysvipc", NULL, 0) 1149 ? JAIL_SYS_INHERIT 1150 : vfs_flagopt(opts, "allow.nosysvipc", NULL, 0) 1151 ? JAIL_SYS_DISABLE 1152 : -1; 1153 if (jsys == JAIL_SYS_DISABLE) { 1154 prison_lock(pr); 1155 orpr = osd_jail_get(pr, shm_prison_slot); 1156 if (orpr != NULL) 1157 osd_jail_del(pr, shm_prison_slot); 1158 prison_unlock(pr); 1159 if (orpr != NULL) { 1160 if (orpr == pr) 1161 shm_prison_cleanup(pr); 1162 /* Disable all child jails as well. */ 1163 FOREACH_PRISON_DESCENDANT(pr, tpr, descend) { 1164 prison_lock(tpr); 1165 trpr = osd_jail_get(tpr, shm_prison_slot); 1166 if (trpr != NULL) { 1167 osd_jail_del(tpr, shm_prison_slot); 1168 prison_unlock(tpr); 1169 if (trpr == tpr) 1170 shm_prison_cleanup(tpr); 1171 } else { 1172 prison_unlock(tpr); 1173 descend = 0; 1174 } 1175 } 1176 } 1177 } else if (jsys != -1) { 1178 if (jsys == JAIL_SYS_NEW) 1179 nrpr = pr; 1180 else { 1181 prison_lock(pr->pr_parent); 1182 nrpr = osd_jail_get(pr->pr_parent, shm_prison_slot); 1183 prison_unlock(pr->pr_parent); 1184 } 1185 rsv = osd_reserve(shm_prison_slot); 1186 prison_lock(pr); 1187 orpr = osd_jail_get(pr, shm_prison_slot); 1188 if (orpr != nrpr) 1189 (void)osd_jail_set_reserved(pr, shm_prison_slot, rsv, 1190 nrpr); 1191 else 1192 osd_free_reserved(rsv); 1193 prison_unlock(pr); 1194 if (orpr != nrpr) { 1195 if (orpr == pr) 1196 shm_prison_cleanup(pr); 1197 if (orpr != NULL) { 1198 /* Change child jails matching the old root, */ 1199 FOREACH_PRISON_DESCENDANT(pr, tpr, descend) { 1200 prison_lock(tpr); 1201 trpr = osd_jail_get(tpr, 1202 shm_prison_slot); 1203 if (trpr == orpr) { 1204 (void)osd_jail_set(tpr, 1205 shm_prison_slot, nrpr); 1206 prison_unlock(tpr); 1207 if (trpr == tpr) 1208 shm_prison_cleanup(tpr); 1209 } else { 1210 prison_unlock(tpr); 1211 descend = 0; 1212 } 1213 } 1214 } 1215 } 1216 } 1217 1218 return (0); 1219 } 1220 1221 static int 1222 shm_prison_get(void *obj, void *data) 1223 { 1224 struct prison *pr = obj; 1225 struct prison *rpr; 1226 struct vfsoptlist *opts = data; 1227 int error, jsys; 1228 1229 /* Set sysvshm based on the jail's root prison. */ 1230 prison_lock(pr); 1231 rpr = osd_jail_get(pr, shm_prison_slot); 1232 prison_unlock(pr); 1233 jsys = rpr == NULL ? JAIL_SYS_DISABLE 1234 : rpr == pr ? JAIL_SYS_NEW : JAIL_SYS_INHERIT; 1235 error = vfs_setopt(opts, "sysvshm", &jsys, sizeof(jsys)); 1236 if (error == ENOENT) 1237 error = 0; 1238 return (error); 1239 } 1240 1241 static int 1242 shm_prison_remove(void *obj, void *data __unused) 1243 { 1244 struct prison *pr = obj; 1245 struct prison *rpr; 1246 1247 SYSVSHM_LOCK(); 1248 prison_lock(pr); 1249 rpr = osd_jail_get(pr, shm_prison_slot); 1250 prison_unlock(pr); 1251 if (rpr == pr) 1252 shm_prison_cleanup(pr); 1253 SYSVSHM_UNLOCK(); 1254 return (0); 1255 } 1256 1257 static void 1258 shm_prison_cleanup(struct prison *pr) 1259 { 1260 struct shmid_kernel *shmseg; 1261 int i; 1262 1263 /* Remove any segments that belong to this jail. */ 1264 for (i = 0; i < shmalloced; i++) { 1265 shmseg = &shmsegs[i]; 1266 if ((shmseg->u.shm_perm.mode & SHMSEG_ALLOCATED) && 1267 shmseg->cred != NULL && shmseg->cred->cr_prison == pr) { 1268 shm_remove(shmseg, i); 1269 } 1270 } 1271 } 1272 1273 SYSCTL_JAIL_PARAM_SYS_NODE(sysvshm, CTLFLAG_RW, "SYSV shared memory"); 1274 1275 #if defined(__i386__) && (defined(COMPAT_FREEBSD4) || defined(COMPAT_43)) 1276 struct oshmid_ds { 1277 struct ipc_perm_old shm_perm; /* operation perms */ 1278 int shm_segsz; /* size of segment (bytes) */ 1279 u_short shm_cpid; /* pid, creator */ 1280 u_short shm_lpid; /* pid, last operation */ 1281 short shm_nattch; /* no. of current attaches */ 1282 time_t shm_atime; /* last attach time */ 1283 time_t shm_dtime; /* last detach time */ 1284 time_t shm_ctime; /* last change time */ 1285 void *shm_handle; /* internal handle for shm segment */ 1286 }; 1287 1288 struct oshmctl_args { 1289 int shmid; 1290 int cmd; 1291 struct oshmid_ds *ubuf; 1292 }; 1293 1294 static int 1295 oshmctl(struct thread *td, struct oshmctl_args *uap) 1296 { 1297 #ifdef COMPAT_43 1298 int error = 0; 1299 struct prison *rpr; 1300 struct shmid_kernel *shmseg; 1301 struct oshmid_ds outbuf; 1302 1303 rpr = shm_find_prison(td->td_ucred); 1304 if (rpr == NULL) 1305 return (ENOSYS); 1306 if (uap->cmd != IPC_STAT) { 1307 return (freebsd7_shmctl(td, 1308 (struct freebsd7_shmctl_args *)uap)); 1309 } 1310 SYSVSHM_LOCK(); 1311 shmseg = shm_find_segment(rpr, uap->shmid, true); 1312 if (shmseg == NULL) { 1313 SYSVSHM_UNLOCK(); 1314 return (EINVAL); 1315 } 1316 error = ipcperm(td, &shmseg->u.shm_perm, IPC_R); 1317 if (error != 0) { 1318 SYSVSHM_UNLOCK(); 1319 return (error); 1320 } 1321 #ifdef MAC 1322 error = mac_sysvshm_check_shmctl(td->td_ucred, shmseg, uap->cmd); 1323 if (error != 0) { 1324 SYSVSHM_UNLOCK(); 1325 return (error); 1326 } 1327 #endif 1328 ipcperm_new2old(&shmseg->u.shm_perm, &outbuf.shm_perm); 1329 outbuf.shm_segsz = shmseg->u.shm_segsz; 1330 outbuf.shm_cpid = shmseg->u.shm_cpid; 1331 outbuf.shm_lpid = shmseg->u.shm_lpid; 1332 outbuf.shm_nattch = shmseg->u.shm_nattch; 1333 outbuf.shm_atime = shmseg->u.shm_atime; 1334 outbuf.shm_dtime = shmseg->u.shm_dtime; 1335 outbuf.shm_ctime = shmseg->u.shm_ctime; 1336 outbuf.shm_handle = shmseg->object; 1337 SYSVSHM_UNLOCK(); 1338 return (copyout(&outbuf, uap->ubuf, sizeof(outbuf))); 1339 #else 1340 return (EINVAL); 1341 #endif 1342 } 1343 1344 /* XXX casting to (sy_call_t *) is bogus, as usual. */ 1345 static sy_call_t *shmcalls[] = { 1346 (sy_call_t *)sys_shmat, (sy_call_t *)oshmctl, 1347 (sy_call_t *)sys_shmdt, (sy_call_t *)sys_shmget, 1348 (sy_call_t *)freebsd7_shmctl 1349 }; 1350 1351 #ifndef _SYS_SYSPROTO_H_ 1352 /* XXX actually varargs. */ 1353 struct shmsys_args { 1354 int which; 1355 int a2; 1356 int a3; 1357 int a4; 1358 }; 1359 #endif 1360 int 1361 sys_shmsys(struct thread *td, struct shmsys_args *uap) 1362 { 1363 1364 AUDIT_ARG_SVIPC_WHICH(uap->which); 1365 if (uap->which < 0 || uap->which >= nitems(shmcalls)) 1366 return (EINVAL); 1367 return ((*shmcalls[uap->which])(td, &uap->a2)); 1368 } 1369 1370 #endif /* i386 && (COMPAT_FREEBSD4 || COMPAT_43) */ 1371 1372 #ifdef COMPAT_FREEBSD32 1373 1374 int 1375 freebsd32_shmsys(struct thread *td, struct freebsd32_shmsys_args *uap) 1376 { 1377 1378 #if defined(COMPAT_FREEBSD4) || defined(COMPAT_FREEBSD5) || \ 1379 defined(COMPAT_FREEBSD6) || defined(COMPAT_FREEBSD7) 1380 AUDIT_ARG_SVIPC_WHICH(uap->which); 1381 switch (uap->which) { 1382 case 0: { /* shmat */ 1383 struct shmat_args ap; 1384 1385 ap.shmid = uap->a2; 1386 ap.shmaddr = PTRIN(uap->a3); 1387 ap.shmflg = uap->a4; 1388 return (sysent[SYS_shmat].sy_call(td, &ap)); 1389 } 1390 case 2: { /* shmdt */ 1391 struct shmdt_args ap; 1392 1393 ap.shmaddr = PTRIN(uap->a2); 1394 return (sysent[SYS_shmdt].sy_call(td, &ap)); 1395 } 1396 case 3: { /* shmget */ 1397 struct shmget_args ap; 1398 1399 ap.key = uap->a2; 1400 ap.size = uap->a3; 1401 ap.shmflg = uap->a4; 1402 return (sysent[SYS_shmget].sy_call(td, &ap)); 1403 } 1404 case 4: { /* shmctl */ 1405 struct freebsd7_freebsd32_shmctl_args ap; 1406 1407 ap.shmid = uap->a2; 1408 ap.cmd = uap->a3; 1409 ap.buf = PTRIN(uap->a4); 1410 return (freebsd7_freebsd32_shmctl(td, &ap)); 1411 } 1412 case 1: /* oshmctl */ 1413 default: 1414 return (EINVAL); 1415 } 1416 #else 1417 return (nosys(td, NULL)); 1418 #endif 1419 } 1420 1421 #if defined(COMPAT_FREEBSD4) || defined(COMPAT_FREEBSD5) || \ 1422 defined(COMPAT_FREEBSD6) || defined(COMPAT_FREEBSD7) 1423 int 1424 freebsd7_freebsd32_shmctl(struct thread *td, 1425 struct freebsd7_freebsd32_shmctl_args *uap) 1426 { 1427 int error; 1428 union { 1429 struct shmid_ds shmid_ds; 1430 struct shm_info shm_info; 1431 struct shminfo shminfo; 1432 } u; 1433 union { 1434 struct shmid_ds_old32 shmid_ds32; 1435 struct shm_info32 shm_info32; 1436 struct shminfo32 shminfo32; 1437 } u32; 1438 size_t sz; 1439 1440 if (uap->cmd == IPC_SET) { 1441 if ((error = copyin(uap->buf, &u32.shmid_ds32, 1442 sizeof(u32.shmid_ds32)))) 1443 goto done; 1444 freebsd32_ipcperm_old_in(&u32.shmid_ds32.shm_perm, 1445 &u.shmid_ds.shm_perm); 1446 CP(u32.shmid_ds32, u.shmid_ds, shm_segsz); 1447 CP(u32.shmid_ds32, u.shmid_ds, shm_lpid); 1448 CP(u32.shmid_ds32, u.shmid_ds, shm_cpid); 1449 CP(u32.shmid_ds32, u.shmid_ds, shm_nattch); 1450 CP(u32.shmid_ds32, u.shmid_ds, shm_atime); 1451 CP(u32.shmid_ds32, u.shmid_ds, shm_dtime); 1452 CP(u32.shmid_ds32, u.shmid_ds, shm_ctime); 1453 } 1454 1455 error = kern_shmctl(td, uap->shmid, uap->cmd, (void *)&u, &sz); 1456 if (error) 1457 goto done; 1458 1459 /* Cases in which we need to copyout */ 1460 switch (uap->cmd) { 1461 case IPC_INFO: 1462 CP(u.shminfo, u32.shminfo32, shmmax); 1463 CP(u.shminfo, u32.shminfo32, shmmin); 1464 CP(u.shminfo, u32.shminfo32, shmmni); 1465 CP(u.shminfo, u32.shminfo32, shmseg); 1466 CP(u.shminfo, u32.shminfo32, shmall); 1467 error = copyout(&u32.shminfo32, uap->buf, 1468 sizeof(u32.shminfo32)); 1469 break; 1470 case SHM_INFO: 1471 CP(u.shm_info, u32.shm_info32, used_ids); 1472 CP(u.shm_info, u32.shm_info32, shm_rss); 1473 CP(u.shm_info, u32.shm_info32, shm_tot); 1474 CP(u.shm_info, u32.shm_info32, shm_swp); 1475 CP(u.shm_info, u32.shm_info32, swap_attempts); 1476 CP(u.shm_info, u32.shm_info32, swap_successes); 1477 error = copyout(&u32.shm_info32, uap->buf, 1478 sizeof(u32.shm_info32)); 1479 break; 1480 case SHM_STAT: 1481 case IPC_STAT: 1482 memset(&u32.shmid_ds32, 0, sizeof(u32.shmid_ds32)); 1483 freebsd32_ipcperm_old_out(&u.shmid_ds.shm_perm, 1484 &u32.shmid_ds32.shm_perm); 1485 if (u.shmid_ds.shm_segsz > INT32_MAX) 1486 u32.shmid_ds32.shm_segsz = INT32_MAX; 1487 else 1488 CP(u.shmid_ds, u32.shmid_ds32, shm_segsz); 1489 CP(u.shmid_ds, u32.shmid_ds32, shm_lpid); 1490 CP(u.shmid_ds, u32.shmid_ds32, shm_cpid); 1491 CP(u.shmid_ds, u32.shmid_ds32, shm_nattch); 1492 CP(u.shmid_ds, u32.shmid_ds32, shm_atime); 1493 CP(u.shmid_ds, u32.shmid_ds32, shm_dtime); 1494 CP(u.shmid_ds, u32.shmid_ds32, shm_ctime); 1495 u32.shmid_ds32.shm_internal = 0; 1496 error = copyout(&u32.shmid_ds32, uap->buf, 1497 sizeof(u32.shmid_ds32)); 1498 break; 1499 } 1500 1501 done: 1502 if (error) { 1503 /* Invalidate the return value */ 1504 td->td_retval[0] = -1; 1505 } 1506 return (error); 1507 } 1508 #endif 1509 1510 int 1511 freebsd32_shmctl(struct thread *td, struct freebsd32_shmctl_args *uap) 1512 { 1513 int error; 1514 union { 1515 struct shmid_ds shmid_ds; 1516 struct shm_info shm_info; 1517 struct shminfo shminfo; 1518 } u; 1519 union { 1520 struct shmid_ds32 shmid_ds32; 1521 struct shm_info32 shm_info32; 1522 struct shminfo32 shminfo32; 1523 } u32; 1524 size_t sz; 1525 1526 if (uap->cmd == IPC_SET) { 1527 if ((error = copyin(uap->buf, &u32.shmid_ds32, 1528 sizeof(u32.shmid_ds32)))) 1529 goto done; 1530 freebsd32_ipcperm_in(&u32.shmid_ds32.shm_perm, 1531 &u.shmid_ds.shm_perm); 1532 CP(u32.shmid_ds32, u.shmid_ds, shm_segsz); 1533 CP(u32.shmid_ds32, u.shmid_ds, shm_lpid); 1534 CP(u32.shmid_ds32, u.shmid_ds, shm_cpid); 1535 CP(u32.shmid_ds32, u.shmid_ds, shm_nattch); 1536 CP(u32.shmid_ds32, u.shmid_ds, shm_atime); 1537 CP(u32.shmid_ds32, u.shmid_ds, shm_dtime); 1538 CP(u32.shmid_ds32, u.shmid_ds, shm_ctime); 1539 } 1540 1541 error = kern_shmctl(td, uap->shmid, uap->cmd, (void *)&u, &sz); 1542 if (error) 1543 goto done; 1544 1545 /* Cases in which we need to copyout */ 1546 switch (uap->cmd) { 1547 case IPC_INFO: 1548 CP(u.shminfo, u32.shminfo32, shmmax); 1549 CP(u.shminfo, u32.shminfo32, shmmin); 1550 CP(u.shminfo, u32.shminfo32, shmmni); 1551 CP(u.shminfo, u32.shminfo32, shmseg); 1552 CP(u.shminfo, u32.shminfo32, shmall); 1553 error = copyout(&u32.shminfo32, uap->buf, 1554 sizeof(u32.shminfo32)); 1555 break; 1556 case SHM_INFO: 1557 CP(u.shm_info, u32.shm_info32, used_ids); 1558 CP(u.shm_info, u32.shm_info32, shm_rss); 1559 CP(u.shm_info, u32.shm_info32, shm_tot); 1560 CP(u.shm_info, u32.shm_info32, shm_swp); 1561 CP(u.shm_info, u32.shm_info32, swap_attempts); 1562 CP(u.shm_info, u32.shm_info32, swap_successes); 1563 error = copyout(&u32.shm_info32, uap->buf, 1564 sizeof(u32.shm_info32)); 1565 break; 1566 case SHM_STAT: 1567 case IPC_STAT: 1568 freebsd32_ipcperm_out(&u.shmid_ds.shm_perm, 1569 &u32.shmid_ds32.shm_perm); 1570 if (u.shmid_ds.shm_segsz > INT32_MAX) 1571 u32.shmid_ds32.shm_segsz = INT32_MAX; 1572 else 1573 CP(u.shmid_ds, u32.shmid_ds32, shm_segsz); 1574 CP(u.shmid_ds, u32.shmid_ds32, shm_lpid); 1575 CP(u.shmid_ds, u32.shmid_ds32, shm_cpid); 1576 CP(u.shmid_ds, u32.shmid_ds32, shm_nattch); 1577 CP(u.shmid_ds, u32.shmid_ds32, shm_atime); 1578 CP(u.shmid_ds, u32.shmid_ds32, shm_dtime); 1579 CP(u.shmid_ds, u32.shmid_ds32, shm_ctime); 1580 error = copyout(&u32.shmid_ds32, uap->buf, 1581 sizeof(u32.shmid_ds32)); 1582 break; 1583 } 1584 1585 done: 1586 if (error) { 1587 /* Invalidate the return value */ 1588 td->td_retval[0] = -1; 1589 } 1590 return (error); 1591 } 1592 #endif 1593 1594 #if defined(COMPAT_FREEBSD4) || defined(COMPAT_FREEBSD5) || \ 1595 defined(COMPAT_FREEBSD6) || defined(COMPAT_FREEBSD7) 1596 1597 #ifndef _SYS_SYSPROTO_H_ 1598 struct freebsd7_shmctl_args { 1599 int shmid; 1600 int cmd; 1601 struct shmid_ds_old *buf; 1602 }; 1603 #endif 1604 int 1605 freebsd7_shmctl(struct thread *td, struct freebsd7_shmctl_args *uap) 1606 { 1607 int error; 1608 struct shmid_ds_old old; 1609 struct shmid_ds buf; 1610 size_t bufsz; 1611 1612 /* 1613 * The only reason IPC_INFO, SHM_INFO, SHM_STAT exists is to support 1614 * Linux binaries. If we see the call come through the FreeBSD ABI, 1615 * return an error back to the user since we do not to support this. 1616 */ 1617 if (uap->cmd == IPC_INFO || uap->cmd == SHM_INFO || 1618 uap->cmd == SHM_STAT) 1619 return (EINVAL); 1620 1621 /* IPC_SET needs to copyin the buffer before calling kern_shmctl */ 1622 if (uap->cmd == IPC_SET) { 1623 if ((error = copyin(uap->buf, &old, sizeof(old)))) 1624 goto done; 1625 ipcperm_old2new(&old.shm_perm, &buf.shm_perm); 1626 CP(old, buf, shm_segsz); 1627 CP(old, buf, shm_lpid); 1628 CP(old, buf, shm_cpid); 1629 CP(old, buf, shm_nattch); 1630 CP(old, buf, shm_atime); 1631 CP(old, buf, shm_dtime); 1632 CP(old, buf, shm_ctime); 1633 } 1634 1635 error = kern_shmctl(td, uap->shmid, uap->cmd, (void *)&buf, &bufsz); 1636 if (error) 1637 goto done; 1638 1639 /* Cases in which we need to copyout */ 1640 switch (uap->cmd) { 1641 case IPC_STAT: 1642 memset(&old, 0, sizeof(old)); 1643 ipcperm_new2old(&buf.shm_perm, &old.shm_perm); 1644 if (buf.shm_segsz > INT_MAX) 1645 old.shm_segsz = INT_MAX; 1646 else 1647 CP(buf, old, shm_segsz); 1648 CP(buf, old, shm_lpid); 1649 CP(buf, old, shm_cpid); 1650 if (buf.shm_nattch > SHRT_MAX) 1651 old.shm_nattch = SHRT_MAX; 1652 else 1653 CP(buf, old, shm_nattch); 1654 CP(buf, old, shm_atime); 1655 CP(buf, old, shm_dtime); 1656 CP(buf, old, shm_ctime); 1657 old.shm_internal = NULL; 1658 error = copyout(&old, uap->buf, sizeof(old)); 1659 break; 1660 } 1661 1662 done: 1663 if (error) { 1664 /* Invalidate the return value */ 1665 td->td_retval[0] = -1; 1666 } 1667 return (error); 1668 } 1669 1670 #endif /* COMPAT_FREEBSD4 || COMPAT_FREEBSD5 || COMPAT_FREEBSD6 || 1671 COMPAT_FREEBSD7 */ 1672 1673 static int 1674 sysvshm_modload(struct module *module, int cmd, void *arg) 1675 { 1676 int error = 0; 1677 1678 switch (cmd) { 1679 case MOD_LOAD: 1680 error = shminit(); 1681 if (error != 0) 1682 shmunload(); 1683 break; 1684 case MOD_UNLOAD: 1685 error = shmunload(); 1686 break; 1687 case MOD_SHUTDOWN: 1688 break; 1689 default: 1690 error = EINVAL; 1691 break; 1692 } 1693 return (error); 1694 } 1695 1696 static moduledata_t sysvshm_mod = { 1697 "sysvshm", 1698 &sysvshm_modload, 1699 NULL 1700 }; 1701 1702 DECLARE_MODULE(sysvshm, sysvshm_mod, SI_SUB_SYSV_SHM, SI_ORDER_FIRST); 1703 MODULE_VERSION(sysvshm, 1); 1704