1 /* $NetBSD: sysv_shm.c,v 1.23 1994/07/04 23:25:12 glass Exp $ */ 2 /*- 3 * Copyright (c) 1994 Adam Glass and Charles Hannum. All rights reserved. 4 * 5 * Redistribution and use in source and binary forms, with or without 6 * modification, are permitted provided that the following conditions 7 * are met: 8 * 1. Redistributions of source code must retain the above copyright 9 * notice, this list of conditions and the following disclaimer. 10 * 2. Redistributions in binary form must reproduce the above copyright 11 * notice, this list of conditions and the following disclaimer in the 12 * documentation and/or other materials provided with the distribution. 13 * 3. All advertising materials mentioning features or use of this software 14 * must display the following acknowledgement: 15 * This product includes software developed by Adam Glass and Charles 16 * Hannum. 17 * 4. The names of the authors may not be used to endorse or promote products 18 * derived from this software without specific prior written permission. 19 * 20 * THIS SOFTWARE IS PROVIDED BY THE AUTHORS ``AS IS'' AND ANY EXPRESS OR 21 * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES 22 * OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. 23 * IN NO EVENT SHALL THE AUTHORS BE LIABLE FOR ANY DIRECT, INDIRECT, 24 * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT 25 * NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, 26 * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY 27 * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT 28 * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF 29 * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. 30 */ 31 /*- 32 * Copyright (c) 2003-2005 McAfee, Inc. 33 * All rights reserved. 34 * 35 * This software was developed for the FreeBSD Project in part by McAfee 36 * Research, the Security Research Division of McAfee, Inc under DARPA/SPAWAR 37 * contract N66001-01-C-8035 ("CBOSS"), as part of the DARPA CHATS research 38 * program. 39 * 40 * Redistribution and use in source and binary forms, with or without 41 * modification, are permitted provided that the following conditions 42 * are met: 43 * 1. Redistributions of source code must retain the above copyright 44 * notice, this list of conditions and the following disclaimer. 45 * 2. Redistributions in binary form must reproduce the above copyright 46 * notice, this list of conditions and the following disclaimer in the 47 * documentation and/or other materials provided with the distribution. 48 * 49 * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND 50 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 51 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 52 * ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE 53 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 54 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 55 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 56 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 57 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 58 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 59 * SUCH DAMAGE. 60 */ 61 62 #include <sys/cdefs.h> 63 __FBSDID("$FreeBSD$"); 64 65 #include "opt_compat.h" 66 #include "opt_sysvipc.h" 67 68 #include <sys/param.h> 69 #include <sys/systm.h> 70 #include <sys/kernel.h> 71 #include <sys/limits.h> 72 #include <sys/lock.h> 73 #include <sys/sysctl.h> 74 #include <sys/shm.h> 75 #include <sys/proc.h> 76 #include <sys/malloc.h> 77 #include <sys/mman.h> 78 #include <sys/module.h> 79 #include <sys/mutex.h> 80 #include <sys/racct.h> 81 #include <sys/resourcevar.h> 82 #include <sys/rwlock.h> 83 #include <sys/stat.h> 84 #include <sys/syscall.h> 85 #include <sys/syscallsubr.h> 86 #include <sys/sysent.h> 87 #include <sys/sysproto.h> 88 #include <sys/jail.h> 89 90 #include <security/mac/mac_framework.h> 91 92 #include <vm/vm.h> 93 #include <vm/vm_param.h> 94 #include <vm/pmap.h> 95 #include <vm/vm_object.h> 96 #include <vm/vm_map.h> 97 #include <vm/vm_page.h> 98 #include <vm/vm_pager.h> 99 100 FEATURE(sysv_shm, "System V shared memory segments support"); 101 102 static MALLOC_DEFINE(M_SHM, "shm", "SVID compatible shared memory segments"); 103 104 static int shmget_allocate_segment(struct thread *td, 105 struct shmget_args *uap, int mode); 106 static int shmget_existing(struct thread *td, struct shmget_args *uap, 107 int mode, int segnum); 108 109 #define SHMSEG_FREE 0x0200 110 #define SHMSEG_REMOVED 0x0400 111 #define SHMSEG_ALLOCATED 0x0800 112 113 static int shm_last_free, shm_nused, shmalloced; 114 vm_size_t shm_committed; 115 static struct shmid_kernel *shmsegs; 116 static unsigned shm_prison_slot; 117 118 struct shmmap_state { 119 vm_offset_t va; 120 int shmid; 121 }; 122 123 static void shm_deallocate_segment(struct shmid_kernel *); 124 static int shm_find_segment_by_key(struct prison *, key_t); 125 static struct shmid_kernel *shm_find_segment(struct prison *, int, bool); 126 static int shm_delete_mapping(struct vmspace *vm, struct shmmap_state *); 127 static void shmrealloc(void); 128 static int shminit(void); 129 static int sysvshm_modload(struct module *, int, void *); 130 static int shmunload(void); 131 static void shmexit_myhook(struct vmspace *vm); 132 static void shmfork_myhook(struct proc *p1, struct proc *p2); 133 static int sysctl_shmsegs(SYSCTL_HANDLER_ARGS); 134 static void shm_remove(struct shmid_kernel *, int); 135 static struct prison *shm_find_prison(struct ucred *); 136 static int shm_prison_cansee(struct prison *, struct shmid_kernel *); 137 static int shm_prison_check(void *, void *); 138 static int shm_prison_set(void *, void *); 139 static int shm_prison_get(void *, void *); 140 static int shm_prison_remove(void *, void *); 141 static void shm_prison_cleanup(struct prison *); 142 143 /* 144 * Tuneable values. 145 */ 146 #ifndef SHMMAXPGS 147 #define SHMMAXPGS 131072 /* Note: sysv shared memory is swap backed. */ 148 #endif 149 #ifndef SHMMAX 150 #define SHMMAX (SHMMAXPGS*PAGE_SIZE) 151 #endif 152 #ifndef SHMMIN 153 #define SHMMIN 1 154 #endif 155 #ifndef SHMMNI 156 #define SHMMNI 192 157 #endif 158 #ifndef SHMSEG 159 #define SHMSEG 128 160 #endif 161 #ifndef SHMALL 162 #define SHMALL (SHMMAXPGS) 163 #endif 164 165 struct shminfo shminfo = { 166 .shmmax = SHMMAX, 167 .shmmin = SHMMIN, 168 .shmmni = SHMMNI, 169 .shmseg = SHMSEG, 170 .shmall = SHMALL 171 }; 172 173 static int shm_use_phys; 174 static int shm_allow_removed = 1; 175 176 SYSCTL_ULONG(_kern_ipc, OID_AUTO, shmmax, CTLFLAG_RWTUN, &shminfo.shmmax, 0, 177 "Maximum shared memory segment size"); 178 SYSCTL_ULONG(_kern_ipc, OID_AUTO, shmmin, CTLFLAG_RWTUN, &shminfo.shmmin, 0, 179 "Minimum shared memory segment size"); 180 SYSCTL_ULONG(_kern_ipc, OID_AUTO, shmmni, CTLFLAG_RDTUN, &shminfo.shmmni, 0, 181 "Number of shared memory identifiers"); 182 SYSCTL_ULONG(_kern_ipc, OID_AUTO, shmseg, CTLFLAG_RDTUN, &shminfo.shmseg, 0, 183 "Number of segments per process"); 184 SYSCTL_ULONG(_kern_ipc, OID_AUTO, shmall, CTLFLAG_RWTUN, &shminfo.shmall, 0, 185 "Maximum number of pages available for shared memory"); 186 SYSCTL_INT(_kern_ipc, OID_AUTO, shm_use_phys, CTLFLAG_RWTUN, 187 &shm_use_phys, 0, "Enable/Disable locking of shared memory pages in core"); 188 SYSCTL_INT(_kern_ipc, OID_AUTO, shm_allow_removed, CTLFLAG_RWTUN, 189 &shm_allow_removed, 0, 190 "Enable/Disable attachment to attached segments marked for removal"); 191 SYSCTL_PROC(_kern_ipc, OID_AUTO, shmsegs, CTLTYPE_OPAQUE | CTLFLAG_RD | 192 CTLFLAG_MPSAFE, NULL, 0, sysctl_shmsegs, "", 193 "Current number of shared memory segments allocated"); 194 195 static struct sx sysvshmsx; 196 #define SYSVSHM_LOCK() sx_xlock(&sysvshmsx) 197 #define SYSVSHM_UNLOCK() sx_xunlock(&sysvshmsx) 198 #define SYSVSHM_ASSERT_LOCKED() sx_assert(&sysvshmsx, SA_XLOCKED) 199 200 static int 201 shm_find_segment_by_key(struct prison *pr, key_t key) 202 { 203 int i; 204 205 for (i = 0; i < shmalloced; i++) 206 if ((shmsegs[i].u.shm_perm.mode & SHMSEG_ALLOCATED) && 207 shmsegs[i].cred != NULL && 208 shmsegs[i].cred->cr_prison == pr && 209 shmsegs[i].u.shm_perm.key == key) 210 return (i); 211 return (-1); 212 } 213 214 /* 215 * Finds segment either by shmid if is_shmid is true, or by segnum if 216 * is_shmid is false. 217 */ 218 static struct shmid_kernel * 219 shm_find_segment(struct prison *rpr, int arg, bool is_shmid) 220 { 221 struct shmid_kernel *shmseg; 222 int segnum; 223 224 segnum = is_shmid ? IPCID_TO_IX(arg) : arg; 225 if (segnum < 0 || segnum >= shmalloced) 226 return (NULL); 227 shmseg = &shmsegs[segnum]; 228 if ((shmseg->u.shm_perm.mode & SHMSEG_ALLOCATED) == 0 || 229 (!shm_allow_removed && 230 (shmseg->u.shm_perm.mode & SHMSEG_REMOVED) != 0) || 231 (is_shmid && shmseg->u.shm_perm.seq != IPCID_TO_SEQ(arg)) || 232 shm_prison_cansee(rpr, shmseg) != 0) 233 return (NULL); 234 return (shmseg); 235 } 236 237 static void 238 shm_deallocate_segment(struct shmid_kernel *shmseg) 239 { 240 vm_size_t size; 241 242 SYSVSHM_ASSERT_LOCKED(); 243 244 vm_object_deallocate(shmseg->object); 245 shmseg->object = NULL; 246 size = round_page(shmseg->u.shm_segsz); 247 shm_committed -= btoc(size); 248 shm_nused--; 249 shmseg->u.shm_perm.mode = SHMSEG_FREE; 250 #ifdef MAC 251 mac_sysvshm_cleanup(shmseg); 252 #endif 253 racct_sub_cred(shmseg->cred, RACCT_NSHM, 1); 254 racct_sub_cred(shmseg->cred, RACCT_SHMSIZE, size); 255 crfree(shmseg->cred); 256 shmseg->cred = NULL; 257 } 258 259 static int 260 shm_delete_mapping(struct vmspace *vm, struct shmmap_state *shmmap_s) 261 { 262 struct shmid_kernel *shmseg; 263 int segnum, result; 264 vm_size_t size; 265 266 SYSVSHM_ASSERT_LOCKED(); 267 segnum = IPCID_TO_IX(shmmap_s->shmid); 268 KASSERT(segnum >= 0 && segnum < shmalloced, 269 ("segnum %d shmalloced %d", segnum, shmalloced)); 270 271 shmseg = &shmsegs[segnum]; 272 size = round_page(shmseg->u.shm_segsz); 273 result = vm_map_remove(&vm->vm_map, shmmap_s->va, shmmap_s->va + size); 274 if (result != KERN_SUCCESS) 275 return (EINVAL); 276 shmmap_s->shmid = -1; 277 shmseg->u.shm_dtime = time_second; 278 if (--shmseg->u.shm_nattch == 0 && 279 (shmseg->u.shm_perm.mode & SHMSEG_REMOVED)) { 280 shm_deallocate_segment(shmseg); 281 shm_last_free = segnum; 282 } 283 return (0); 284 } 285 286 static void 287 shm_remove(struct shmid_kernel *shmseg, int segnum) 288 { 289 290 shmseg->u.shm_perm.key = IPC_PRIVATE; 291 shmseg->u.shm_perm.mode |= SHMSEG_REMOVED; 292 if (shmseg->u.shm_nattch == 0) { 293 shm_deallocate_segment(shmseg); 294 shm_last_free = segnum; 295 } 296 } 297 298 static struct prison * 299 shm_find_prison(struct ucred *cred) 300 { 301 struct prison *pr, *rpr; 302 303 pr = cred->cr_prison; 304 prison_lock(pr); 305 rpr = osd_jail_get(pr, shm_prison_slot); 306 prison_unlock(pr); 307 return rpr; 308 } 309 310 static int 311 shm_prison_cansee(struct prison *rpr, struct shmid_kernel *shmseg) 312 { 313 314 if (shmseg->cred == NULL || 315 !(rpr == shmseg->cred->cr_prison || 316 prison_ischild(rpr, shmseg->cred->cr_prison))) 317 return (EINVAL); 318 return (0); 319 } 320 321 static int 322 kern_shmdt_locked(struct thread *td, const void *shmaddr) 323 { 324 struct proc *p = td->td_proc; 325 struct shmmap_state *shmmap_s; 326 #ifdef MAC 327 struct shmid_kernel *shmsegptr; 328 int error; 329 #endif 330 int i; 331 332 SYSVSHM_ASSERT_LOCKED(); 333 if (shm_find_prison(td->td_ucred) == NULL) 334 return (ENOSYS); 335 shmmap_s = p->p_vmspace->vm_shm; 336 if (shmmap_s == NULL) 337 return (EINVAL); 338 for (i = 0; i < shminfo.shmseg; i++, shmmap_s++) { 339 if (shmmap_s->shmid != -1 && 340 shmmap_s->va == (vm_offset_t)shmaddr) { 341 break; 342 } 343 } 344 if (i == shminfo.shmseg) 345 return (EINVAL); 346 #ifdef MAC 347 shmsegptr = &shmsegs[IPCID_TO_IX(shmmap_s->shmid)]; 348 error = mac_sysvshm_check_shmdt(td->td_ucred, shmsegptr); 349 if (error != 0) 350 return (error); 351 #endif 352 return (shm_delete_mapping(p->p_vmspace, shmmap_s)); 353 } 354 355 #ifndef _SYS_SYSPROTO_H_ 356 struct shmdt_args { 357 const void *shmaddr; 358 }; 359 #endif 360 int 361 sys_shmdt(struct thread *td, struct shmdt_args *uap) 362 { 363 int error; 364 365 SYSVSHM_LOCK(); 366 error = kern_shmdt_locked(td, uap->shmaddr); 367 SYSVSHM_UNLOCK(); 368 return (error); 369 } 370 371 static int 372 kern_shmat_locked(struct thread *td, int shmid, const void *shmaddr, 373 int shmflg) 374 { 375 struct prison *rpr; 376 struct proc *p = td->td_proc; 377 struct shmid_kernel *shmseg; 378 struct shmmap_state *shmmap_s; 379 vm_offset_t attach_va; 380 vm_prot_t prot; 381 vm_size_t size; 382 int error, i, rv; 383 384 SYSVSHM_ASSERT_LOCKED(); 385 rpr = shm_find_prison(td->td_ucred); 386 if (rpr == NULL) 387 return (ENOSYS); 388 shmmap_s = p->p_vmspace->vm_shm; 389 if (shmmap_s == NULL) { 390 shmmap_s = malloc(shminfo.shmseg * sizeof(struct shmmap_state), 391 M_SHM, M_WAITOK); 392 for (i = 0; i < shminfo.shmseg; i++) 393 shmmap_s[i].shmid = -1; 394 KASSERT(p->p_vmspace->vm_shm == NULL, ("raced")); 395 p->p_vmspace->vm_shm = shmmap_s; 396 } 397 shmseg = shm_find_segment(rpr, shmid, true); 398 if (shmseg == NULL) 399 return (EINVAL); 400 error = ipcperm(td, &shmseg->u.shm_perm, 401 (shmflg & SHM_RDONLY) ? IPC_R : IPC_R|IPC_W); 402 if (error != 0) 403 return (error); 404 #ifdef MAC 405 error = mac_sysvshm_check_shmat(td->td_ucred, shmseg, shmflg); 406 if (error != 0) 407 return (error); 408 #endif 409 for (i = 0; i < shminfo.shmseg; i++) { 410 if (shmmap_s->shmid == -1) 411 break; 412 shmmap_s++; 413 } 414 if (i >= shminfo.shmseg) 415 return (EMFILE); 416 size = round_page(shmseg->u.shm_segsz); 417 prot = VM_PROT_READ; 418 if ((shmflg & SHM_RDONLY) == 0) 419 prot |= VM_PROT_WRITE; 420 if (shmaddr != NULL) { 421 if ((shmflg & SHM_RND) != 0) 422 attach_va = rounddown2((vm_offset_t)shmaddr, SHMLBA); 423 else if (((vm_offset_t)shmaddr & (SHMLBA-1)) == 0) 424 attach_va = (vm_offset_t)shmaddr; 425 else 426 return (EINVAL); 427 } else { 428 /* 429 * This is just a hint to vm_map_find() about where to 430 * put it. 431 */ 432 attach_va = round_page((vm_offset_t)p->p_vmspace->vm_daddr + 433 lim_max(td, RLIMIT_DATA)); 434 } 435 436 vm_object_reference(shmseg->object); 437 rv = vm_map_find(&p->p_vmspace->vm_map, shmseg->object, 0, &attach_va, 438 size, 0, shmaddr != NULL ? VMFS_NO_SPACE : VMFS_OPTIMAL_SPACE, 439 prot, prot, MAP_INHERIT_SHARE | MAP_PREFAULT_PARTIAL); 440 if (rv != KERN_SUCCESS) { 441 vm_object_deallocate(shmseg->object); 442 return (ENOMEM); 443 } 444 445 shmmap_s->va = attach_va; 446 shmmap_s->shmid = shmid; 447 shmseg->u.shm_lpid = p->p_pid; 448 shmseg->u.shm_atime = time_second; 449 shmseg->u.shm_nattch++; 450 td->td_retval[0] = attach_va; 451 return (error); 452 } 453 454 int 455 kern_shmat(struct thread *td, int shmid, const void *shmaddr, int shmflg) 456 { 457 int error; 458 459 SYSVSHM_LOCK(); 460 error = kern_shmat_locked(td, shmid, shmaddr, shmflg); 461 SYSVSHM_UNLOCK(); 462 return (error); 463 } 464 465 #ifndef _SYS_SYSPROTO_H_ 466 struct shmat_args { 467 int shmid; 468 const void *shmaddr; 469 int shmflg; 470 }; 471 #endif 472 int 473 sys_shmat(struct thread *td, struct shmat_args *uap) 474 { 475 476 return (kern_shmat(td, uap->shmid, uap->shmaddr, uap->shmflg)); 477 } 478 479 static int 480 kern_shmctl_locked(struct thread *td, int shmid, int cmd, void *buf, 481 size_t *bufsz) 482 { 483 struct prison *rpr; 484 struct shmid_kernel *shmseg; 485 struct shmid_ds *shmidp; 486 struct shm_info shm_info; 487 int error; 488 489 SYSVSHM_ASSERT_LOCKED(); 490 491 rpr = shm_find_prison(td->td_ucred); 492 if (rpr == NULL) 493 return (ENOSYS); 494 495 switch (cmd) { 496 /* 497 * It is possible that kern_shmctl is being called from the Linux ABI 498 * layer, in which case, we will need to implement IPC_INFO. It should 499 * be noted that other shmctl calls will be funneled through here for 500 * Linix binaries as well. 501 * 502 * NB: The Linux ABI layer will convert this data to structure(s) more 503 * consistent with the Linux ABI. 504 */ 505 case IPC_INFO: 506 memcpy(buf, &shminfo, sizeof(shminfo)); 507 if (bufsz) 508 *bufsz = sizeof(shminfo); 509 td->td_retval[0] = shmalloced; 510 return (0); 511 case SHM_INFO: { 512 shm_info.used_ids = shm_nused; 513 shm_info.shm_rss = 0; /*XXX where to get from ? */ 514 shm_info.shm_tot = 0; /*XXX where to get from ? */ 515 shm_info.shm_swp = 0; /*XXX where to get from ? */ 516 shm_info.swap_attempts = 0; /*XXX where to get from ? */ 517 shm_info.swap_successes = 0; /*XXX where to get from ? */ 518 memcpy(buf, &shm_info, sizeof(shm_info)); 519 if (bufsz != NULL) 520 *bufsz = sizeof(shm_info); 521 td->td_retval[0] = shmalloced; 522 return (0); 523 } 524 } 525 shmseg = shm_find_segment(rpr, shmid, cmd != SHM_STAT); 526 if (shmseg == NULL) 527 return (EINVAL); 528 #ifdef MAC 529 error = mac_sysvshm_check_shmctl(td->td_ucred, shmseg, cmd); 530 if (error != 0) 531 return (error); 532 #endif 533 switch (cmd) { 534 case SHM_STAT: 535 case IPC_STAT: 536 shmidp = (struct shmid_ds *)buf; 537 error = ipcperm(td, &shmseg->u.shm_perm, IPC_R); 538 if (error != 0) 539 return (error); 540 memcpy(shmidp, &shmseg->u, sizeof(struct shmid_ds)); 541 if (td->td_ucred->cr_prison != shmseg->cred->cr_prison) 542 shmidp->shm_perm.key = IPC_PRIVATE; 543 if (bufsz != NULL) 544 *bufsz = sizeof(struct shmid_ds); 545 if (cmd == SHM_STAT) { 546 td->td_retval[0] = IXSEQ_TO_IPCID(shmid, 547 shmseg->u.shm_perm); 548 } 549 break; 550 case IPC_SET: 551 shmidp = (struct shmid_ds *)buf; 552 error = ipcperm(td, &shmseg->u.shm_perm, IPC_M); 553 if (error != 0) 554 return (error); 555 shmseg->u.shm_perm.uid = shmidp->shm_perm.uid; 556 shmseg->u.shm_perm.gid = shmidp->shm_perm.gid; 557 shmseg->u.shm_perm.mode = 558 (shmseg->u.shm_perm.mode & ~ACCESSPERMS) | 559 (shmidp->shm_perm.mode & ACCESSPERMS); 560 shmseg->u.shm_ctime = time_second; 561 break; 562 case IPC_RMID: 563 error = ipcperm(td, &shmseg->u.shm_perm, IPC_M); 564 if (error != 0) 565 return (error); 566 shm_remove(shmseg, IPCID_TO_IX(shmid)); 567 break; 568 #if 0 569 case SHM_LOCK: 570 case SHM_UNLOCK: 571 #endif 572 default: 573 error = EINVAL; 574 break; 575 } 576 return (error); 577 } 578 579 int 580 kern_shmctl(struct thread *td, int shmid, int cmd, void *buf, size_t *bufsz) 581 { 582 int error; 583 584 SYSVSHM_LOCK(); 585 error = kern_shmctl_locked(td, shmid, cmd, buf, bufsz); 586 SYSVSHM_UNLOCK(); 587 return (error); 588 } 589 590 591 #ifndef _SYS_SYSPROTO_H_ 592 struct shmctl_args { 593 int shmid; 594 int cmd; 595 struct shmid_ds *buf; 596 }; 597 #endif 598 int 599 sys_shmctl(struct thread *td, struct shmctl_args *uap) 600 { 601 int error; 602 struct shmid_ds buf; 603 size_t bufsz; 604 605 /* 606 * The only reason IPC_INFO, SHM_INFO, SHM_STAT exists is to support 607 * Linux binaries. If we see the call come through the FreeBSD ABI, 608 * return an error back to the user since we do not to support this. 609 */ 610 if (uap->cmd == IPC_INFO || uap->cmd == SHM_INFO || 611 uap->cmd == SHM_STAT) 612 return (EINVAL); 613 614 /* IPC_SET needs to copyin the buffer before calling kern_shmctl */ 615 if (uap->cmd == IPC_SET) { 616 if ((error = copyin(uap->buf, &buf, sizeof(struct shmid_ds)))) 617 goto done; 618 } 619 620 error = kern_shmctl(td, uap->shmid, uap->cmd, (void *)&buf, &bufsz); 621 if (error) 622 goto done; 623 624 /* Cases in which we need to copyout */ 625 switch (uap->cmd) { 626 case IPC_STAT: 627 error = copyout(&buf, uap->buf, bufsz); 628 break; 629 } 630 631 done: 632 if (error) { 633 /* Invalidate the return value */ 634 td->td_retval[0] = -1; 635 } 636 return (error); 637 } 638 639 640 static int 641 shmget_existing(struct thread *td, struct shmget_args *uap, int mode, 642 int segnum) 643 { 644 struct shmid_kernel *shmseg; 645 #ifdef MAC 646 int error; 647 #endif 648 649 SYSVSHM_ASSERT_LOCKED(); 650 KASSERT(segnum >= 0 && segnum < shmalloced, 651 ("segnum %d shmalloced %d", segnum, shmalloced)); 652 shmseg = &shmsegs[segnum]; 653 if ((uap->shmflg & (IPC_CREAT | IPC_EXCL)) == (IPC_CREAT | IPC_EXCL)) 654 return (EEXIST); 655 #ifdef MAC 656 error = mac_sysvshm_check_shmget(td->td_ucred, shmseg, uap->shmflg); 657 if (error != 0) 658 return (error); 659 #endif 660 if (uap->size != 0 && uap->size > shmseg->u.shm_segsz) 661 return (EINVAL); 662 td->td_retval[0] = IXSEQ_TO_IPCID(segnum, shmseg->u.shm_perm); 663 return (0); 664 } 665 666 static int 667 shmget_allocate_segment(struct thread *td, struct shmget_args *uap, int mode) 668 { 669 struct ucred *cred = td->td_ucred; 670 struct shmid_kernel *shmseg; 671 vm_object_t shm_object; 672 int i, segnum; 673 size_t size; 674 675 SYSVSHM_ASSERT_LOCKED(); 676 677 if (uap->size < shminfo.shmmin || uap->size > shminfo.shmmax) 678 return (EINVAL); 679 if (shm_nused >= shminfo.shmmni) /* Any shmids left? */ 680 return (ENOSPC); 681 size = round_page(uap->size); 682 if (shm_committed + btoc(size) > shminfo.shmall) 683 return (ENOMEM); 684 if (shm_last_free < 0) { 685 shmrealloc(); /* Maybe expand the shmsegs[] array. */ 686 for (i = 0; i < shmalloced; i++) 687 if (shmsegs[i].u.shm_perm.mode & SHMSEG_FREE) 688 break; 689 if (i == shmalloced) 690 return (ENOSPC); 691 segnum = i; 692 } else { 693 segnum = shm_last_free; 694 shm_last_free = -1; 695 } 696 KASSERT(segnum >= 0 && segnum < shmalloced, 697 ("segnum %d shmalloced %d", segnum, shmalloced)); 698 shmseg = &shmsegs[segnum]; 699 #ifdef RACCT 700 if (racct_enable) { 701 PROC_LOCK(td->td_proc); 702 if (racct_add(td->td_proc, RACCT_NSHM, 1)) { 703 PROC_UNLOCK(td->td_proc); 704 return (ENOSPC); 705 } 706 if (racct_add(td->td_proc, RACCT_SHMSIZE, size)) { 707 racct_sub(td->td_proc, RACCT_NSHM, 1); 708 PROC_UNLOCK(td->td_proc); 709 return (ENOMEM); 710 } 711 PROC_UNLOCK(td->td_proc); 712 } 713 #endif 714 715 /* 716 * We make sure that we have allocated a pager before we need 717 * to. 718 */ 719 shm_object = vm_pager_allocate(shm_use_phys ? OBJT_PHYS : OBJT_SWAP, 720 0, size, VM_PROT_DEFAULT, 0, cred); 721 if (shm_object == NULL) { 722 #ifdef RACCT 723 if (racct_enable) { 724 PROC_LOCK(td->td_proc); 725 racct_sub(td->td_proc, RACCT_NSHM, 1); 726 racct_sub(td->td_proc, RACCT_SHMSIZE, size); 727 PROC_UNLOCK(td->td_proc); 728 } 729 #endif 730 return (ENOMEM); 731 } 732 shm_object->pg_color = 0; 733 VM_OBJECT_WLOCK(shm_object); 734 vm_object_clear_flag(shm_object, OBJ_ONEMAPPING); 735 vm_object_set_flag(shm_object, OBJ_COLORED | OBJ_NOSPLIT); 736 VM_OBJECT_WUNLOCK(shm_object); 737 738 shmseg->object = shm_object; 739 shmseg->u.shm_perm.cuid = shmseg->u.shm_perm.uid = cred->cr_uid; 740 shmseg->u.shm_perm.cgid = shmseg->u.shm_perm.gid = cred->cr_gid; 741 shmseg->u.shm_perm.mode = (mode & ACCESSPERMS) | SHMSEG_ALLOCATED; 742 shmseg->u.shm_perm.key = uap->key; 743 shmseg->u.shm_perm.seq = (shmseg->u.shm_perm.seq + 1) & 0x7fff; 744 shmseg->cred = crhold(cred); 745 shmseg->u.shm_segsz = uap->size; 746 shmseg->u.shm_cpid = td->td_proc->p_pid; 747 shmseg->u.shm_lpid = shmseg->u.shm_nattch = 0; 748 shmseg->u.shm_atime = shmseg->u.shm_dtime = 0; 749 #ifdef MAC 750 mac_sysvshm_create(cred, shmseg); 751 #endif 752 shmseg->u.shm_ctime = time_second; 753 shm_committed += btoc(size); 754 shm_nused++; 755 td->td_retval[0] = IXSEQ_TO_IPCID(segnum, shmseg->u.shm_perm); 756 757 return (0); 758 } 759 760 #ifndef _SYS_SYSPROTO_H_ 761 struct shmget_args { 762 key_t key; 763 size_t size; 764 int shmflg; 765 }; 766 #endif 767 int 768 sys_shmget(struct thread *td, struct shmget_args *uap) 769 { 770 int segnum, mode; 771 int error; 772 773 if (shm_find_prison(td->td_ucred) == NULL) 774 return (ENOSYS); 775 mode = uap->shmflg & ACCESSPERMS; 776 SYSVSHM_LOCK(); 777 if (uap->key == IPC_PRIVATE) { 778 error = shmget_allocate_segment(td, uap, mode); 779 } else { 780 segnum = shm_find_segment_by_key(td->td_ucred->cr_prison, 781 uap->key); 782 if (segnum >= 0) 783 error = shmget_existing(td, uap, mode, segnum); 784 else if ((uap->shmflg & IPC_CREAT) == 0) 785 error = ENOENT; 786 else 787 error = shmget_allocate_segment(td, uap, mode); 788 } 789 SYSVSHM_UNLOCK(); 790 return (error); 791 } 792 793 static void 794 shmfork_myhook(struct proc *p1, struct proc *p2) 795 { 796 struct shmmap_state *shmmap_s; 797 size_t size; 798 int i; 799 800 SYSVSHM_LOCK(); 801 size = shminfo.shmseg * sizeof(struct shmmap_state); 802 shmmap_s = malloc(size, M_SHM, M_WAITOK); 803 bcopy(p1->p_vmspace->vm_shm, shmmap_s, size); 804 p2->p_vmspace->vm_shm = shmmap_s; 805 for (i = 0; i < shminfo.shmseg; i++, shmmap_s++) { 806 if (shmmap_s->shmid != -1) { 807 KASSERT(IPCID_TO_IX(shmmap_s->shmid) >= 0 && 808 IPCID_TO_IX(shmmap_s->shmid) < shmalloced, 809 ("segnum %d shmalloced %d", 810 IPCID_TO_IX(shmmap_s->shmid), shmalloced)); 811 shmsegs[IPCID_TO_IX(shmmap_s->shmid)].u.shm_nattch++; 812 } 813 } 814 SYSVSHM_UNLOCK(); 815 } 816 817 static void 818 shmexit_myhook(struct vmspace *vm) 819 { 820 struct shmmap_state *base, *shm; 821 int i; 822 823 base = vm->vm_shm; 824 if (base != NULL) { 825 vm->vm_shm = NULL; 826 SYSVSHM_LOCK(); 827 for (i = 0, shm = base; i < shminfo.shmseg; i++, shm++) { 828 if (shm->shmid != -1) 829 shm_delete_mapping(vm, shm); 830 } 831 SYSVSHM_UNLOCK(); 832 free(base, M_SHM); 833 } 834 } 835 836 static void 837 shmrealloc(void) 838 { 839 struct shmid_kernel *newsegs; 840 int i; 841 842 SYSVSHM_ASSERT_LOCKED(); 843 844 if (shmalloced >= shminfo.shmmni) 845 return; 846 847 newsegs = malloc(shminfo.shmmni * sizeof(*newsegs), M_SHM, M_WAITOK); 848 for (i = 0; i < shmalloced; i++) 849 bcopy(&shmsegs[i], &newsegs[i], sizeof(newsegs[0])); 850 for (; i < shminfo.shmmni; i++) { 851 newsegs[i].u.shm_perm.mode = SHMSEG_FREE; 852 newsegs[i].u.shm_perm.seq = 0; 853 #ifdef MAC 854 mac_sysvshm_init(&newsegs[i]); 855 #endif 856 } 857 free(shmsegs, M_SHM); 858 shmsegs = newsegs; 859 shmalloced = shminfo.shmmni; 860 } 861 862 static struct syscall_helper_data shm_syscalls[] = { 863 SYSCALL_INIT_HELPER(shmat), 864 SYSCALL_INIT_HELPER(shmctl), 865 SYSCALL_INIT_HELPER(shmdt), 866 SYSCALL_INIT_HELPER(shmget), 867 #if defined(COMPAT_FREEBSD4) || defined(COMPAT_FREEBSD5) || \ 868 defined(COMPAT_FREEBSD6) || defined(COMPAT_FREEBSD7) 869 SYSCALL_INIT_HELPER_COMPAT(freebsd7_shmctl), 870 #endif 871 #if defined(__i386__) && (defined(COMPAT_FREEBSD4) || defined(COMPAT_43)) 872 SYSCALL_INIT_HELPER(shmsys), 873 #endif 874 SYSCALL_INIT_LAST 875 }; 876 877 #ifdef COMPAT_FREEBSD32 878 #include <compat/freebsd32/freebsd32.h> 879 #include <compat/freebsd32/freebsd32_ipc.h> 880 #include <compat/freebsd32/freebsd32_proto.h> 881 #include <compat/freebsd32/freebsd32_signal.h> 882 #include <compat/freebsd32/freebsd32_syscall.h> 883 #include <compat/freebsd32/freebsd32_util.h> 884 885 static struct syscall_helper_data shm32_syscalls[] = { 886 SYSCALL32_INIT_HELPER_COMPAT(shmat), 887 SYSCALL32_INIT_HELPER_COMPAT(shmdt), 888 SYSCALL32_INIT_HELPER_COMPAT(shmget), 889 SYSCALL32_INIT_HELPER(freebsd32_shmsys), 890 SYSCALL32_INIT_HELPER(freebsd32_shmctl), 891 #if defined(COMPAT_FREEBSD4) || defined(COMPAT_FREEBSD5) || \ 892 defined(COMPAT_FREEBSD6) || defined(COMPAT_FREEBSD7) 893 SYSCALL32_INIT_HELPER(freebsd7_freebsd32_shmctl), 894 #endif 895 SYSCALL_INIT_LAST 896 }; 897 #endif 898 899 static int 900 shminit(void) 901 { 902 struct prison *pr; 903 void **rsv; 904 int i, error; 905 osd_method_t methods[PR_MAXMETHOD] = { 906 [PR_METHOD_CHECK] = shm_prison_check, 907 [PR_METHOD_SET] = shm_prison_set, 908 [PR_METHOD_GET] = shm_prison_get, 909 [PR_METHOD_REMOVE] = shm_prison_remove, 910 }; 911 912 #ifndef BURN_BRIDGES 913 if (TUNABLE_ULONG_FETCH("kern.ipc.shmmaxpgs", &shminfo.shmall) != 0) 914 printf("kern.ipc.shmmaxpgs is now called kern.ipc.shmall!\n"); 915 #endif 916 if (shminfo.shmmax == SHMMAX) { 917 /* Initialize shmmax dealing with possible overflow. */ 918 for (i = PAGE_SIZE; i != 0; i--) { 919 shminfo.shmmax = shminfo.shmall * i; 920 if ((shminfo.shmmax / shminfo.shmall) == (u_long)i) 921 break; 922 } 923 } 924 shmalloced = shminfo.shmmni; 925 shmsegs = malloc(shmalloced * sizeof(shmsegs[0]), M_SHM, M_WAITOK); 926 for (i = 0; i < shmalloced; i++) { 927 shmsegs[i].u.shm_perm.mode = SHMSEG_FREE; 928 shmsegs[i].u.shm_perm.seq = 0; 929 #ifdef MAC 930 mac_sysvshm_init(&shmsegs[i]); 931 #endif 932 } 933 shm_last_free = 0; 934 shm_nused = 0; 935 shm_committed = 0; 936 sx_init(&sysvshmsx, "sysvshmsx"); 937 shmexit_hook = &shmexit_myhook; 938 shmfork_hook = &shmfork_myhook; 939 940 /* Set current prisons according to their allow.sysvipc. */ 941 shm_prison_slot = osd_jail_register(NULL, methods); 942 rsv = osd_reserve(shm_prison_slot); 943 prison_lock(&prison0); 944 (void)osd_jail_set_reserved(&prison0, shm_prison_slot, rsv, &prison0); 945 prison_unlock(&prison0); 946 rsv = NULL; 947 sx_slock(&allprison_lock); 948 TAILQ_FOREACH(pr, &allprison, pr_list) { 949 if (rsv == NULL) 950 rsv = osd_reserve(shm_prison_slot); 951 prison_lock(pr); 952 if ((pr->pr_allow & PR_ALLOW_SYSVIPC) && pr->pr_ref > 0) { 953 (void)osd_jail_set_reserved(pr, shm_prison_slot, rsv, 954 &prison0); 955 rsv = NULL; 956 } 957 prison_unlock(pr); 958 } 959 if (rsv != NULL) 960 osd_free_reserved(rsv); 961 sx_sunlock(&allprison_lock); 962 963 error = syscall_helper_register(shm_syscalls, SY_THR_STATIC_KLD); 964 if (error != 0) 965 return (error); 966 #ifdef COMPAT_FREEBSD32 967 error = syscall32_helper_register(shm32_syscalls, SY_THR_STATIC_KLD); 968 if (error != 0) 969 return (error); 970 #endif 971 return (0); 972 } 973 974 static int 975 shmunload(void) 976 { 977 int i; 978 979 if (shm_nused > 0) 980 return (EBUSY); 981 982 #ifdef COMPAT_FREEBSD32 983 syscall32_helper_unregister(shm32_syscalls); 984 #endif 985 syscall_helper_unregister(shm_syscalls); 986 if (shm_prison_slot != 0) 987 osd_jail_deregister(shm_prison_slot); 988 989 for (i = 0; i < shmalloced; i++) { 990 #ifdef MAC 991 mac_sysvshm_destroy(&shmsegs[i]); 992 #endif 993 /* 994 * Objects might be still mapped into the processes 995 * address spaces. Actual free would happen on the 996 * last mapping destruction. 997 */ 998 if (shmsegs[i].u.shm_perm.mode != SHMSEG_FREE) 999 vm_object_deallocate(shmsegs[i].object); 1000 } 1001 free(shmsegs, M_SHM); 1002 shmexit_hook = NULL; 1003 shmfork_hook = NULL; 1004 sx_destroy(&sysvshmsx); 1005 return (0); 1006 } 1007 1008 static int 1009 sysctl_shmsegs(SYSCTL_HANDLER_ARGS) 1010 { 1011 struct shmid_kernel tshmseg; 1012 struct prison *pr, *rpr; 1013 int error, i; 1014 1015 SYSVSHM_LOCK(); 1016 pr = req->td->td_ucred->cr_prison; 1017 rpr = shm_find_prison(req->td->td_ucred); 1018 error = 0; 1019 for (i = 0; i < shmalloced; i++) { 1020 if ((shmsegs[i].u.shm_perm.mode & SHMSEG_ALLOCATED) == 0 || 1021 rpr == NULL || shm_prison_cansee(rpr, &shmsegs[i]) != 0) { 1022 bzero(&tshmseg, sizeof(tshmseg)); 1023 tshmseg.u.shm_perm.mode = SHMSEG_FREE; 1024 } else { 1025 tshmseg = shmsegs[i]; 1026 if (tshmseg.cred->cr_prison != pr) 1027 tshmseg.u.shm_perm.key = IPC_PRIVATE; 1028 } 1029 error = SYSCTL_OUT(req, &tshmseg, sizeof(tshmseg)); 1030 if (error != 0) 1031 break; 1032 } 1033 SYSVSHM_UNLOCK(); 1034 return (error); 1035 } 1036 1037 static int 1038 shm_prison_check(void *obj, void *data) 1039 { 1040 struct prison *pr = obj; 1041 struct prison *prpr; 1042 struct vfsoptlist *opts = data; 1043 int error, jsys; 1044 1045 /* 1046 * sysvshm is a jailsys integer. 1047 * It must be "disable" if the parent jail is disabled. 1048 */ 1049 error = vfs_copyopt(opts, "sysvshm", &jsys, sizeof(jsys)); 1050 if (error != ENOENT) { 1051 if (error != 0) 1052 return (error); 1053 switch (jsys) { 1054 case JAIL_SYS_DISABLE: 1055 break; 1056 case JAIL_SYS_NEW: 1057 case JAIL_SYS_INHERIT: 1058 prison_lock(pr->pr_parent); 1059 prpr = osd_jail_get(pr->pr_parent, shm_prison_slot); 1060 prison_unlock(pr->pr_parent); 1061 if (prpr == NULL) 1062 return (EPERM); 1063 break; 1064 default: 1065 return (EINVAL); 1066 } 1067 } 1068 1069 return (0); 1070 } 1071 1072 static int 1073 shm_prison_set(void *obj, void *data) 1074 { 1075 struct prison *pr = obj; 1076 struct prison *tpr, *orpr, *nrpr, *trpr; 1077 struct vfsoptlist *opts = data; 1078 void *rsv; 1079 int jsys, descend; 1080 1081 /* 1082 * sysvshm controls which jail is the root of the associated segments 1083 * (this jail or same as the parent), or if the feature is available 1084 * at all. 1085 */ 1086 if (vfs_copyopt(opts, "sysvshm", &jsys, sizeof(jsys)) == ENOENT) 1087 jsys = vfs_flagopt(opts, "allow.sysvipc", NULL, 0) 1088 ? JAIL_SYS_INHERIT 1089 : vfs_flagopt(opts, "allow.nosysvipc", NULL, 0) 1090 ? JAIL_SYS_DISABLE 1091 : -1; 1092 if (jsys == JAIL_SYS_DISABLE) { 1093 prison_lock(pr); 1094 orpr = osd_jail_get(pr, shm_prison_slot); 1095 if (orpr != NULL) 1096 osd_jail_del(pr, shm_prison_slot); 1097 prison_unlock(pr); 1098 if (orpr != NULL) { 1099 if (orpr == pr) 1100 shm_prison_cleanup(pr); 1101 /* Disable all child jails as well. */ 1102 FOREACH_PRISON_DESCENDANT(pr, tpr, descend) { 1103 prison_lock(tpr); 1104 trpr = osd_jail_get(tpr, shm_prison_slot); 1105 if (trpr != NULL) { 1106 osd_jail_del(tpr, shm_prison_slot); 1107 prison_unlock(tpr); 1108 if (trpr == tpr) 1109 shm_prison_cleanup(tpr); 1110 } else { 1111 prison_unlock(tpr); 1112 descend = 0; 1113 } 1114 } 1115 } 1116 } else if (jsys != -1) { 1117 if (jsys == JAIL_SYS_NEW) 1118 nrpr = pr; 1119 else { 1120 prison_lock(pr->pr_parent); 1121 nrpr = osd_jail_get(pr->pr_parent, shm_prison_slot); 1122 prison_unlock(pr->pr_parent); 1123 } 1124 rsv = osd_reserve(shm_prison_slot); 1125 prison_lock(pr); 1126 orpr = osd_jail_get(pr, shm_prison_slot); 1127 if (orpr != nrpr) 1128 (void)osd_jail_set_reserved(pr, shm_prison_slot, rsv, 1129 nrpr); 1130 else 1131 osd_free_reserved(rsv); 1132 prison_unlock(pr); 1133 if (orpr != nrpr) { 1134 if (orpr == pr) 1135 shm_prison_cleanup(pr); 1136 if (orpr != NULL) { 1137 /* Change child jails matching the old root, */ 1138 FOREACH_PRISON_DESCENDANT(pr, tpr, descend) { 1139 prison_lock(tpr); 1140 trpr = osd_jail_get(tpr, 1141 shm_prison_slot); 1142 if (trpr == orpr) { 1143 (void)osd_jail_set(tpr, 1144 shm_prison_slot, nrpr); 1145 prison_unlock(tpr); 1146 if (trpr == tpr) 1147 shm_prison_cleanup(tpr); 1148 } else { 1149 prison_unlock(tpr); 1150 descend = 0; 1151 } 1152 } 1153 } 1154 } 1155 } 1156 1157 return (0); 1158 } 1159 1160 static int 1161 shm_prison_get(void *obj, void *data) 1162 { 1163 struct prison *pr = obj; 1164 struct prison *rpr; 1165 struct vfsoptlist *opts = data; 1166 int error, jsys; 1167 1168 /* Set sysvshm based on the jail's root prison. */ 1169 prison_lock(pr); 1170 rpr = osd_jail_get(pr, shm_prison_slot); 1171 prison_unlock(pr); 1172 jsys = rpr == NULL ? JAIL_SYS_DISABLE 1173 : rpr == pr ? JAIL_SYS_NEW : JAIL_SYS_INHERIT; 1174 error = vfs_setopt(opts, "sysvshm", &jsys, sizeof(jsys)); 1175 if (error == ENOENT) 1176 error = 0; 1177 return (error); 1178 } 1179 1180 static int 1181 shm_prison_remove(void *obj, void *data __unused) 1182 { 1183 struct prison *pr = obj; 1184 struct prison *rpr; 1185 1186 SYSVSHM_LOCK(); 1187 prison_lock(pr); 1188 rpr = osd_jail_get(pr, shm_prison_slot); 1189 prison_unlock(pr); 1190 if (rpr == pr) 1191 shm_prison_cleanup(pr); 1192 SYSVSHM_UNLOCK(); 1193 return (0); 1194 } 1195 1196 static void 1197 shm_prison_cleanup(struct prison *pr) 1198 { 1199 struct shmid_kernel *shmseg; 1200 int i; 1201 1202 /* Remove any segments that belong to this jail. */ 1203 for (i = 0; i < shmalloced; i++) { 1204 shmseg = &shmsegs[i]; 1205 if ((shmseg->u.shm_perm.mode & SHMSEG_ALLOCATED) && 1206 shmseg->cred != NULL && shmseg->cred->cr_prison == pr) { 1207 shm_remove(shmseg, i); 1208 } 1209 } 1210 } 1211 1212 SYSCTL_JAIL_PARAM_SYS_NODE(sysvshm, CTLFLAG_RW, "SYSV shared memory"); 1213 1214 #if defined(__i386__) && (defined(COMPAT_FREEBSD4) || defined(COMPAT_43)) 1215 struct oshmid_ds { 1216 struct ipc_perm_old shm_perm; /* operation perms */ 1217 int shm_segsz; /* size of segment (bytes) */ 1218 u_short shm_cpid; /* pid, creator */ 1219 u_short shm_lpid; /* pid, last operation */ 1220 short shm_nattch; /* no. of current attaches */ 1221 time_t shm_atime; /* last attach time */ 1222 time_t shm_dtime; /* last detach time */ 1223 time_t shm_ctime; /* last change time */ 1224 void *shm_handle; /* internal handle for shm segment */ 1225 }; 1226 1227 struct oshmctl_args { 1228 int shmid; 1229 int cmd; 1230 struct oshmid_ds *ubuf; 1231 }; 1232 1233 static int 1234 oshmctl(struct thread *td, struct oshmctl_args *uap) 1235 { 1236 #ifdef COMPAT_43 1237 int error = 0; 1238 struct prison *rpr; 1239 struct shmid_kernel *shmseg; 1240 struct oshmid_ds outbuf; 1241 1242 rpr = shm_find_prison(td->td_ucred); 1243 if (rpr == NULL) 1244 return (ENOSYS); 1245 if (uap->cmd != IPC_STAT) { 1246 return (freebsd7_shmctl(td, 1247 (struct freebsd7_shmctl_args *)uap)); 1248 } 1249 SYSVSHM_LOCK(); 1250 shmseg = shm_find_segment(rpr, uap->shmid, true); 1251 if (shmseg == NULL) { 1252 SYSVSHM_UNLOCK(); 1253 return (EINVAL); 1254 } 1255 error = ipcperm(td, &shmseg->u.shm_perm, IPC_R); 1256 if (error != 0) { 1257 SYSVSHM_UNLOCK(); 1258 return (error); 1259 } 1260 #ifdef MAC 1261 error = mac_sysvshm_check_shmctl(td->td_ucred, shmseg, uap->cmd); 1262 if (error != 0) { 1263 SYSVSHM_UNLOCK(); 1264 return (error); 1265 } 1266 #endif 1267 ipcperm_new2old(&shmseg->u.shm_perm, &outbuf.shm_perm); 1268 outbuf.shm_segsz = shmseg->u.shm_segsz; 1269 outbuf.shm_cpid = shmseg->u.shm_cpid; 1270 outbuf.shm_lpid = shmseg->u.shm_lpid; 1271 outbuf.shm_nattch = shmseg->u.shm_nattch; 1272 outbuf.shm_atime = shmseg->u.shm_atime; 1273 outbuf.shm_dtime = shmseg->u.shm_dtime; 1274 outbuf.shm_ctime = shmseg->u.shm_ctime; 1275 outbuf.shm_handle = shmseg->object; 1276 SYSVSHM_UNLOCK(); 1277 return (copyout(&outbuf, uap->ubuf, sizeof(outbuf))); 1278 #else 1279 return (EINVAL); 1280 #endif 1281 } 1282 1283 /* XXX casting to (sy_call_t *) is bogus, as usual. */ 1284 static sy_call_t *shmcalls[] = { 1285 (sy_call_t *)sys_shmat, (sy_call_t *)oshmctl, 1286 (sy_call_t *)sys_shmdt, (sy_call_t *)sys_shmget, 1287 (sy_call_t *)freebsd7_shmctl 1288 }; 1289 1290 #ifndef _SYS_SYSPROTO_H_ 1291 /* XXX actually varargs. */ 1292 struct shmsys_args { 1293 int which; 1294 int a2; 1295 int a3; 1296 int a4; 1297 }; 1298 #endif 1299 int 1300 sys_shmsys(struct thread *td, struct shmsys_args *uap) 1301 { 1302 1303 if (uap->which < 0 || uap->which >= nitems(shmcalls)) 1304 return (EINVAL); 1305 return ((*shmcalls[uap->which])(td, &uap->a2)); 1306 } 1307 1308 #endif /* i386 && (COMPAT_FREEBSD4 || COMPAT_43) */ 1309 1310 #ifdef COMPAT_FREEBSD32 1311 1312 int 1313 freebsd32_shmsys(struct thread *td, struct freebsd32_shmsys_args *uap) 1314 { 1315 1316 #if defined(COMPAT_FREEBSD4) || defined(COMPAT_FREEBSD5) || \ 1317 defined(COMPAT_FREEBSD6) || defined(COMPAT_FREEBSD7) 1318 switch (uap->which) { 1319 case 0: { /* shmat */ 1320 struct shmat_args ap; 1321 1322 ap.shmid = uap->a2; 1323 ap.shmaddr = PTRIN(uap->a3); 1324 ap.shmflg = uap->a4; 1325 return (sysent[SYS_shmat].sy_call(td, &ap)); 1326 } 1327 case 2: { /* shmdt */ 1328 struct shmdt_args ap; 1329 1330 ap.shmaddr = PTRIN(uap->a2); 1331 return (sysent[SYS_shmdt].sy_call(td, &ap)); 1332 } 1333 case 3: { /* shmget */ 1334 struct shmget_args ap; 1335 1336 ap.key = uap->a2; 1337 ap.size = uap->a3; 1338 ap.shmflg = uap->a4; 1339 return (sysent[SYS_shmget].sy_call(td, &ap)); 1340 } 1341 case 4: { /* shmctl */ 1342 struct freebsd7_freebsd32_shmctl_args ap; 1343 1344 ap.shmid = uap->a2; 1345 ap.cmd = uap->a3; 1346 ap.buf = PTRIN(uap->a4); 1347 return (freebsd7_freebsd32_shmctl(td, &ap)); 1348 } 1349 case 1: /* oshmctl */ 1350 default: 1351 return (EINVAL); 1352 } 1353 #else 1354 return (nosys(td, NULL)); 1355 #endif 1356 } 1357 1358 #if defined(COMPAT_FREEBSD4) || defined(COMPAT_FREEBSD5) || \ 1359 defined(COMPAT_FREEBSD6) || defined(COMPAT_FREEBSD7) 1360 int 1361 freebsd7_freebsd32_shmctl(struct thread *td, 1362 struct freebsd7_freebsd32_shmctl_args *uap) 1363 { 1364 int error; 1365 union { 1366 struct shmid_ds shmid_ds; 1367 struct shm_info shm_info; 1368 struct shminfo shminfo; 1369 } u; 1370 union { 1371 struct shmid_ds32_old shmid_ds32; 1372 struct shm_info32 shm_info32; 1373 struct shminfo32 shminfo32; 1374 } u32; 1375 size_t sz; 1376 1377 if (uap->cmd == IPC_SET) { 1378 if ((error = copyin(uap->buf, &u32.shmid_ds32, 1379 sizeof(u32.shmid_ds32)))) 1380 goto done; 1381 freebsd32_ipcperm_old_in(&u32.shmid_ds32.shm_perm, 1382 &u.shmid_ds.shm_perm); 1383 CP(u32.shmid_ds32, u.shmid_ds, shm_segsz); 1384 CP(u32.shmid_ds32, u.shmid_ds, shm_lpid); 1385 CP(u32.shmid_ds32, u.shmid_ds, shm_cpid); 1386 CP(u32.shmid_ds32, u.shmid_ds, shm_nattch); 1387 CP(u32.shmid_ds32, u.shmid_ds, shm_atime); 1388 CP(u32.shmid_ds32, u.shmid_ds, shm_dtime); 1389 CP(u32.shmid_ds32, u.shmid_ds, shm_ctime); 1390 } 1391 1392 error = kern_shmctl(td, uap->shmid, uap->cmd, (void *)&u, &sz); 1393 if (error) 1394 goto done; 1395 1396 /* Cases in which we need to copyout */ 1397 switch (uap->cmd) { 1398 case IPC_INFO: 1399 CP(u.shminfo, u32.shminfo32, shmmax); 1400 CP(u.shminfo, u32.shminfo32, shmmin); 1401 CP(u.shminfo, u32.shminfo32, shmmni); 1402 CP(u.shminfo, u32.shminfo32, shmseg); 1403 CP(u.shminfo, u32.shminfo32, shmall); 1404 error = copyout(&u32.shminfo32, uap->buf, 1405 sizeof(u32.shminfo32)); 1406 break; 1407 case SHM_INFO: 1408 CP(u.shm_info, u32.shm_info32, used_ids); 1409 CP(u.shm_info, u32.shm_info32, shm_rss); 1410 CP(u.shm_info, u32.shm_info32, shm_tot); 1411 CP(u.shm_info, u32.shm_info32, shm_swp); 1412 CP(u.shm_info, u32.shm_info32, swap_attempts); 1413 CP(u.shm_info, u32.shm_info32, swap_successes); 1414 error = copyout(&u32.shm_info32, uap->buf, 1415 sizeof(u32.shm_info32)); 1416 break; 1417 case SHM_STAT: 1418 case IPC_STAT: 1419 freebsd32_ipcperm_old_out(&u.shmid_ds.shm_perm, 1420 &u32.shmid_ds32.shm_perm); 1421 if (u.shmid_ds.shm_segsz > INT32_MAX) 1422 u32.shmid_ds32.shm_segsz = INT32_MAX; 1423 else 1424 CP(u.shmid_ds, u32.shmid_ds32, shm_segsz); 1425 CP(u.shmid_ds, u32.shmid_ds32, shm_lpid); 1426 CP(u.shmid_ds, u32.shmid_ds32, shm_cpid); 1427 CP(u.shmid_ds, u32.shmid_ds32, shm_nattch); 1428 CP(u.shmid_ds, u32.shmid_ds32, shm_atime); 1429 CP(u.shmid_ds, u32.shmid_ds32, shm_dtime); 1430 CP(u.shmid_ds, u32.shmid_ds32, shm_ctime); 1431 u32.shmid_ds32.shm_internal = 0; 1432 error = copyout(&u32.shmid_ds32, uap->buf, 1433 sizeof(u32.shmid_ds32)); 1434 break; 1435 } 1436 1437 done: 1438 if (error) { 1439 /* Invalidate the return value */ 1440 td->td_retval[0] = -1; 1441 } 1442 return (error); 1443 } 1444 #endif 1445 1446 int 1447 freebsd32_shmctl(struct thread *td, struct freebsd32_shmctl_args *uap) 1448 { 1449 int error; 1450 union { 1451 struct shmid_ds shmid_ds; 1452 struct shm_info shm_info; 1453 struct shminfo shminfo; 1454 } u; 1455 union { 1456 struct shmid_ds32 shmid_ds32; 1457 struct shm_info32 shm_info32; 1458 struct shminfo32 shminfo32; 1459 } u32; 1460 size_t sz; 1461 1462 if (uap->cmd == IPC_SET) { 1463 if ((error = copyin(uap->buf, &u32.shmid_ds32, 1464 sizeof(u32.shmid_ds32)))) 1465 goto done; 1466 freebsd32_ipcperm_in(&u32.shmid_ds32.shm_perm, 1467 &u.shmid_ds.shm_perm); 1468 CP(u32.shmid_ds32, u.shmid_ds, shm_segsz); 1469 CP(u32.shmid_ds32, u.shmid_ds, shm_lpid); 1470 CP(u32.shmid_ds32, u.shmid_ds, shm_cpid); 1471 CP(u32.shmid_ds32, u.shmid_ds, shm_nattch); 1472 CP(u32.shmid_ds32, u.shmid_ds, shm_atime); 1473 CP(u32.shmid_ds32, u.shmid_ds, shm_dtime); 1474 CP(u32.shmid_ds32, u.shmid_ds, shm_ctime); 1475 } 1476 1477 error = kern_shmctl(td, uap->shmid, uap->cmd, (void *)&u, &sz); 1478 if (error) 1479 goto done; 1480 1481 /* Cases in which we need to copyout */ 1482 switch (uap->cmd) { 1483 case IPC_INFO: 1484 CP(u.shminfo, u32.shminfo32, shmmax); 1485 CP(u.shminfo, u32.shminfo32, shmmin); 1486 CP(u.shminfo, u32.shminfo32, shmmni); 1487 CP(u.shminfo, u32.shminfo32, shmseg); 1488 CP(u.shminfo, u32.shminfo32, shmall); 1489 error = copyout(&u32.shminfo32, uap->buf, 1490 sizeof(u32.shminfo32)); 1491 break; 1492 case SHM_INFO: 1493 CP(u.shm_info, u32.shm_info32, used_ids); 1494 CP(u.shm_info, u32.shm_info32, shm_rss); 1495 CP(u.shm_info, u32.shm_info32, shm_tot); 1496 CP(u.shm_info, u32.shm_info32, shm_swp); 1497 CP(u.shm_info, u32.shm_info32, swap_attempts); 1498 CP(u.shm_info, u32.shm_info32, swap_successes); 1499 error = copyout(&u32.shm_info32, uap->buf, 1500 sizeof(u32.shm_info32)); 1501 break; 1502 case SHM_STAT: 1503 case IPC_STAT: 1504 freebsd32_ipcperm_out(&u.shmid_ds.shm_perm, 1505 &u32.shmid_ds32.shm_perm); 1506 if (u.shmid_ds.shm_segsz > INT32_MAX) 1507 u32.shmid_ds32.shm_segsz = INT32_MAX; 1508 else 1509 CP(u.shmid_ds, u32.shmid_ds32, shm_segsz); 1510 CP(u.shmid_ds, u32.shmid_ds32, shm_lpid); 1511 CP(u.shmid_ds, u32.shmid_ds32, shm_cpid); 1512 CP(u.shmid_ds, u32.shmid_ds32, shm_nattch); 1513 CP(u.shmid_ds, u32.shmid_ds32, shm_atime); 1514 CP(u.shmid_ds, u32.shmid_ds32, shm_dtime); 1515 CP(u.shmid_ds, u32.shmid_ds32, shm_ctime); 1516 error = copyout(&u32.shmid_ds32, uap->buf, 1517 sizeof(u32.shmid_ds32)); 1518 break; 1519 } 1520 1521 done: 1522 if (error) { 1523 /* Invalidate the return value */ 1524 td->td_retval[0] = -1; 1525 } 1526 return (error); 1527 } 1528 #endif 1529 1530 #if defined(COMPAT_FREEBSD4) || defined(COMPAT_FREEBSD5) || \ 1531 defined(COMPAT_FREEBSD6) || defined(COMPAT_FREEBSD7) 1532 1533 #ifndef CP 1534 #define CP(src, dst, fld) do { (dst).fld = (src).fld; } while (0) 1535 #endif 1536 1537 #ifndef _SYS_SYSPROTO_H_ 1538 struct freebsd7_shmctl_args { 1539 int shmid; 1540 int cmd; 1541 struct shmid_ds_old *buf; 1542 }; 1543 #endif 1544 int 1545 freebsd7_shmctl(struct thread *td, struct freebsd7_shmctl_args *uap) 1546 { 1547 int error; 1548 struct shmid_ds_old old; 1549 struct shmid_ds buf; 1550 size_t bufsz; 1551 1552 /* 1553 * The only reason IPC_INFO, SHM_INFO, SHM_STAT exists is to support 1554 * Linux binaries. If we see the call come through the FreeBSD ABI, 1555 * return an error back to the user since we do not to support this. 1556 */ 1557 if (uap->cmd == IPC_INFO || uap->cmd == SHM_INFO || 1558 uap->cmd == SHM_STAT) 1559 return (EINVAL); 1560 1561 /* IPC_SET needs to copyin the buffer before calling kern_shmctl */ 1562 if (uap->cmd == IPC_SET) { 1563 if ((error = copyin(uap->buf, &old, sizeof(old)))) 1564 goto done; 1565 ipcperm_old2new(&old.shm_perm, &buf.shm_perm); 1566 CP(old, buf, shm_segsz); 1567 CP(old, buf, shm_lpid); 1568 CP(old, buf, shm_cpid); 1569 CP(old, buf, shm_nattch); 1570 CP(old, buf, shm_atime); 1571 CP(old, buf, shm_dtime); 1572 CP(old, buf, shm_ctime); 1573 } 1574 1575 error = kern_shmctl(td, uap->shmid, uap->cmd, (void *)&buf, &bufsz); 1576 if (error) 1577 goto done; 1578 1579 /* Cases in which we need to copyout */ 1580 switch (uap->cmd) { 1581 case IPC_STAT: 1582 ipcperm_new2old(&buf.shm_perm, &old.shm_perm); 1583 if (buf.shm_segsz > INT_MAX) 1584 old.shm_segsz = INT_MAX; 1585 else 1586 CP(buf, old, shm_segsz); 1587 CP(buf, old, shm_lpid); 1588 CP(buf, old, shm_cpid); 1589 if (buf.shm_nattch > SHRT_MAX) 1590 old.shm_nattch = SHRT_MAX; 1591 else 1592 CP(buf, old, shm_nattch); 1593 CP(buf, old, shm_atime); 1594 CP(buf, old, shm_dtime); 1595 CP(buf, old, shm_ctime); 1596 old.shm_internal = NULL; 1597 error = copyout(&old, uap->buf, sizeof(old)); 1598 break; 1599 } 1600 1601 done: 1602 if (error) { 1603 /* Invalidate the return value */ 1604 td->td_retval[0] = -1; 1605 } 1606 return (error); 1607 } 1608 1609 #endif /* COMPAT_FREEBSD4 || COMPAT_FREEBSD5 || COMPAT_FREEBSD6 || 1610 COMPAT_FREEBSD7 */ 1611 1612 static int 1613 sysvshm_modload(struct module *module, int cmd, void *arg) 1614 { 1615 int error = 0; 1616 1617 switch (cmd) { 1618 case MOD_LOAD: 1619 error = shminit(); 1620 if (error != 0) 1621 shmunload(); 1622 break; 1623 case MOD_UNLOAD: 1624 error = shmunload(); 1625 break; 1626 case MOD_SHUTDOWN: 1627 break; 1628 default: 1629 error = EINVAL; 1630 break; 1631 } 1632 return (error); 1633 } 1634 1635 static moduledata_t sysvshm_mod = { 1636 "sysvshm", 1637 &sysvshm_modload, 1638 NULL 1639 }; 1640 1641 DECLARE_MODULE(sysvshm, sysvshm_mod, SI_SUB_SYSV_SHM, SI_ORDER_FIRST); 1642 MODULE_VERSION(sysvshm, 1); 1643