1 /* $NetBSD: kern_proc.c,v 1.180 2011/05/13 22:22:03 rmind Exp $ */ 2 3 /*- 4 * Copyright (c) 1999, 2006, 2007, 2008 The NetBSD Foundation, Inc. 5 * All rights reserved. 6 * 7 * This code is derived from software contributed to The NetBSD Foundation 8 * by Jason R. Thorpe of the Numerical Aerospace Simulation Facility, 9 * NASA Ames Research Center, and by Andrew Doran. 10 * 11 * Redistribution and use in source and binary forms, with or without 12 * modification, are permitted provided that the following conditions 13 * are met: 14 * 1. Redistributions of source code must retain the above copyright 15 * notice, this list of conditions and the following disclaimer. 16 * 2. Redistributions in binary form must reproduce the above copyright 17 * notice, this list of conditions and the following disclaimer in the 18 * documentation and/or other materials provided with the distribution. 19 * 20 * THIS SOFTWARE IS PROVIDED BY THE NETBSD FOUNDATION, INC. AND CONTRIBUTORS 21 * ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED 22 * TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR 23 * PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE FOUNDATION OR CONTRIBUTORS 24 * BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR 25 * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF 26 * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS 27 * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN 28 * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) 29 * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE 30 * POSSIBILITY OF SUCH DAMAGE. 31 */ 32 33 /* 34 * Copyright (c) 1982, 1986, 1989, 1991, 1993 35 * The Regents of the University of California. All rights reserved. 36 * 37 * Redistribution and use in source and binary forms, with or without 38 * modification, are permitted provided that the following conditions 39 * are met: 40 * 1. Redistributions of source code must retain the above copyright 41 * notice, this list of conditions and the following disclaimer. 42 * 2. Redistributions in binary form must reproduce the above copyright 43 * notice, this list of conditions and the following disclaimer in the 44 * documentation and/or other materials provided with the distribution. 45 * 3. Neither the name of the University nor the names of its contributors 46 * may be used to endorse or promote products derived from this software 47 * without specific prior written permission. 48 * 49 * THIS SOFTWARE IS PROVIDED BY THE REGENTS 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 REGENTS 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 * @(#)kern_proc.c 8.7 (Berkeley) 2/14/95 62 */ 63 64 #include <sys/cdefs.h> 65 __KERNEL_RCSID(0, "$NetBSD: kern_proc.c,v 1.180 2011/05/13 22:22:03 rmind Exp $"); 66 67 #ifdef _KERNEL_OPT 68 #include "opt_kstack.h" 69 #include "opt_maxuprc.h" 70 #include "opt_dtrace.h" 71 #include "opt_compat_netbsd32.h" 72 #endif 73 74 #include <sys/param.h> 75 #include <sys/systm.h> 76 #include <sys/kernel.h> 77 #include <sys/proc.h> 78 #include <sys/resourcevar.h> 79 #include <sys/buf.h> 80 #include <sys/acct.h> 81 #include <sys/wait.h> 82 #include <sys/file.h> 83 #include <ufs/ufs/quota.h> 84 #include <sys/uio.h> 85 #include <sys/pool.h> 86 #include <sys/pset.h> 87 #include <sys/mbuf.h> 88 #include <sys/ioctl.h> 89 #include <sys/tty.h> 90 #include <sys/signalvar.h> 91 #include <sys/ras.h> 92 #include <sys/sa.h> 93 #include <sys/savar.h> 94 #include <sys/filedesc.h> 95 #include "sys/syscall_stats.h" 96 #include <sys/kauth.h> 97 #include <sys/sleepq.h> 98 #include <sys/atomic.h> 99 #include <sys/kmem.h> 100 #include <sys/dtrace_bsd.h> 101 #include <sys/sysctl.h> 102 #include <sys/exec.h> 103 #include <sys/cpu.h> 104 105 #include <uvm/uvm_extern.h> 106 #include <uvm/uvm_extern.h> 107 108 #ifdef COMPAT_NETBSD32 109 #include <compat/netbsd32/netbsd32.h> 110 #endif 111 112 /* 113 * Process lists. 114 */ 115 116 struct proclist allproc __cacheline_aligned; 117 struct proclist zombproc __cacheline_aligned; 118 119 kmutex_t * proc_lock __cacheline_aligned; 120 121 /* 122 * pid to proc lookup is done by indexing the pid_table array. 123 * Since pid numbers are only allocated when an empty slot 124 * has been found, there is no need to search any lists ever. 125 * (an orphaned pgrp will lock the slot, a session will lock 126 * the pgrp with the same number.) 127 * If the table is too small it is reallocated with twice the 128 * previous size and the entries 'unzipped' into the two halves. 129 * A linked list of free entries is passed through the pt_proc 130 * field of 'free' items - set odd to be an invalid ptr. 131 */ 132 133 struct pid_table { 134 struct proc *pt_proc; 135 struct pgrp *pt_pgrp; 136 pid_t pt_pid; 137 }; 138 #if 1 /* strongly typed cast - should be a noop */ 139 static inline uint p2u(struct proc *p) { return (uint)(uintptr_t)p; } 140 #else 141 #define p2u(p) ((uint)p) 142 #endif 143 #define P_VALID(p) (!(p2u(p) & 1)) 144 #define P_NEXT(p) (p2u(p) >> 1) 145 #define P_FREE(pid) ((struct proc *)(uintptr_t)((pid) << 1 | 1)) 146 147 /* 148 * Table of process IDs (PIDs). 149 */ 150 static struct pid_table *pid_table __read_mostly; 151 152 #define INITIAL_PID_TABLE_SIZE (1 << 5) 153 154 /* Table mask, threshold for growing and number of allocated PIDs. */ 155 static u_int pid_tbl_mask __read_mostly; 156 static u_int pid_alloc_lim __read_mostly; 157 static u_int pid_alloc_cnt __cacheline_aligned; 158 159 /* Next free, last free and maximum PIDs. */ 160 static u_int next_free_pt __cacheline_aligned; 161 static u_int last_free_pt __cacheline_aligned; 162 static pid_t pid_max __read_mostly; 163 164 /* Components of the first process -- never freed. */ 165 166 extern struct emul emul_netbsd; /* defined in kern_exec.c */ 167 168 struct session session0 = { 169 .s_count = 1, 170 .s_sid = 0, 171 }; 172 struct pgrp pgrp0 = { 173 .pg_members = LIST_HEAD_INITIALIZER(&pgrp0.pg_members), 174 .pg_session = &session0, 175 }; 176 filedesc_t filedesc0; 177 struct cwdinfo cwdi0 = { 178 .cwdi_cmask = CMASK, /* see cmask below */ 179 .cwdi_refcnt = 1, 180 }; 181 struct plimit limit0; 182 struct pstats pstat0; 183 struct vmspace vmspace0; 184 struct sigacts sigacts0; 185 struct proc proc0 = { 186 .p_lwps = LIST_HEAD_INITIALIZER(&proc0.p_lwps), 187 .p_sigwaiters = LIST_HEAD_INITIALIZER(&proc0.p_sigwaiters), 188 .p_nlwps = 1, 189 .p_nrlwps = 1, 190 .p_nlwpid = 1, /* must match lwp0.l_lid */ 191 .p_pgrp = &pgrp0, 192 .p_comm = "system", 193 /* 194 * Set P_NOCLDWAIT so that kernel threads are reparented to init(8) 195 * when they exit. init(8) can easily wait them out for us. 196 */ 197 .p_flag = PK_SYSTEM | PK_NOCLDWAIT, 198 .p_stat = SACTIVE, 199 .p_nice = NZERO, 200 .p_emul = &emul_netbsd, 201 .p_cwdi = &cwdi0, 202 .p_limit = &limit0, 203 .p_fd = &filedesc0, 204 .p_vmspace = &vmspace0, 205 .p_stats = &pstat0, 206 .p_sigacts = &sigacts0, 207 }; 208 kauth_cred_t cred0; 209 210 static const int nofile = NOFILE; 211 static const int maxuprc = MAXUPRC; 212 static const int cmask = CMASK; 213 214 static int sysctl_doeproc(SYSCTLFN_PROTO); 215 static int sysctl_kern_proc_args(SYSCTLFN_PROTO); 216 static void fill_kproc2(struct proc *, struct kinfo_proc2 *, bool); 217 218 /* 219 * The process list descriptors, used during pid allocation and 220 * by sysctl. No locking on this data structure is needed since 221 * it is completely static. 222 */ 223 const struct proclist_desc proclists[] = { 224 { &allproc }, 225 { &zombproc }, 226 { NULL }, 227 }; 228 229 static struct pgrp * pg_remove(pid_t); 230 static void pg_delete(pid_t); 231 static void orphanpg(struct pgrp *); 232 233 static specificdata_domain_t proc_specificdata_domain; 234 235 static pool_cache_t proc_cache; 236 237 static kauth_listener_t proc_listener; 238 239 static int 240 proc_listener_cb(kauth_cred_t cred, kauth_action_t action, void *cookie, 241 void *arg0, void *arg1, void *arg2, void *arg3) 242 { 243 struct proc *p; 244 int result; 245 246 result = KAUTH_RESULT_DEFER; 247 p = arg0; 248 249 switch (action) { 250 case KAUTH_PROCESS_CANSEE: { 251 enum kauth_process_req req; 252 253 req = (enum kauth_process_req)arg1; 254 255 switch (req) { 256 case KAUTH_REQ_PROCESS_CANSEE_ARGS: 257 case KAUTH_REQ_PROCESS_CANSEE_ENTRY: 258 case KAUTH_REQ_PROCESS_CANSEE_OPENFILES: 259 result = KAUTH_RESULT_ALLOW; 260 261 break; 262 263 case KAUTH_REQ_PROCESS_CANSEE_ENV: 264 if (kauth_cred_getuid(cred) != 265 kauth_cred_getuid(p->p_cred) || 266 kauth_cred_getuid(cred) != 267 kauth_cred_getsvuid(p->p_cred)) 268 break; 269 270 result = KAUTH_RESULT_ALLOW; 271 272 break; 273 274 default: 275 break; 276 } 277 278 break; 279 } 280 281 case KAUTH_PROCESS_FORK: { 282 int lnprocs = (int)(unsigned long)arg2; 283 284 /* 285 * Don't allow a nonprivileged user to use the last few 286 * processes. The variable lnprocs is the current number of 287 * processes, maxproc is the limit. 288 */ 289 if (__predict_false((lnprocs >= maxproc - 5))) 290 break; 291 292 result = KAUTH_RESULT_ALLOW; 293 294 break; 295 } 296 297 case KAUTH_PROCESS_CORENAME: 298 case KAUTH_PROCESS_STOPFLAG: 299 if (proc_uidmatch(cred, p->p_cred) == 0) 300 result = KAUTH_RESULT_ALLOW; 301 302 break; 303 304 default: 305 break; 306 } 307 308 return result; 309 } 310 311 /* 312 * Initialize global process hashing structures. 313 */ 314 void 315 procinit(void) 316 { 317 const struct proclist_desc *pd; 318 u_int i; 319 #define LINK_EMPTY ((PID_MAX + INITIAL_PID_TABLE_SIZE) & ~(INITIAL_PID_TABLE_SIZE - 1)) 320 321 for (pd = proclists; pd->pd_list != NULL; pd++) 322 LIST_INIT(pd->pd_list); 323 324 proc_lock = mutex_obj_alloc(MUTEX_DEFAULT, IPL_NONE); 325 pid_table = kmem_alloc(INITIAL_PID_TABLE_SIZE 326 * sizeof(struct pid_table), KM_SLEEP); 327 pid_tbl_mask = INITIAL_PID_TABLE_SIZE - 1; 328 pid_max = PID_MAX; 329 330 /* Set free list running through table... 331 Preset 'use count' above PID_MAX so we allocate pid 1 next. */ 332 for (i = 0; i <= pid_tbl_mask; i++) { 333 pid_table[i].pt_proc = P_FREE(LINK_EMPTY + i + 1); 334 pid_table[i].pt_pgrp = 0; 335 pid_table[i].pt_pid = 0; 336 } 337 /* slot 0 is just grabbed */ 338 next_free_pt = 1; 339 /* Need to fix last entry. */ 340 last_free_pt = pid_tbl_mask; 341 pid_table[last_free_pt].pt_proc = P_FREE(LINK_EMPTY); 342 /* point at which we grow table - to avoid reusing pids too often */ 343 pid_alloc_lim = pid_tbl_mask - 1; 344 #undef LINK_EMPTY 345 346 proc_specificdata_domain = specificdata_domain_create(); 347 KASSERT(proc_specificdata_domain != NULL); 348 349 proc_cache = pool_cache_init(sizeof(struct proc), 0, 0, 0, 350 "procpl", NULL, IPL_NONE, NULL, NULL, NULL); 351 352 proc_listener = kauth_listen_scope(KAUTH_SCOPE_PROCESS, 353 proc_listener_cb, NULL); 354 } 355 356 void 357 procinit_sysctl(void) 358 { 359 static struct sysctllog *clog; 360 361 sysctl_createv(&clog, 0, NULL, NULL, 362 CTLFLAG_PERMANENT, 363 CTLTYPE_NODE, "kern", NULL, 364 NULL, 0, NULL, 0, 365 CTL_KERN, CTL_EOL); 366 367 sysctl_createv(&clog, 0, NULL, NULL, 368 CTLFLAG_PERMANENT, 369 CTLTYPE_NODE, "proc", 370 SYSCTL_DESCR("System-wide process information"), 371 sysctl_doeproc, 0, NULL, 0, 372 CTL_KERN, KERN_PROC, CTL_EOL); 373 sysctl_createv(&clog, 0, NULL, NULL, 374 CTLFLAG_PERMANENT, 375 CTLTYPE_NODE, "proc2", 376 SYSCTL_DESCR("Machine-independent process information"), 377 sysctl_doeproc, 0, NULL, 0, 378 CTL_KERN, KERN_PROC2, CTL_EOL); 379 sysctl_createv(&clog, 0, NULL, NULL, 380 CTLFLAG_PERMANENT, 381 CTLTYPE_NODE, "proc_args", 382 SYSCTL_DESCR("Process argument information"), 383 sysctl_kern_proc_args, 0, NULL, 0, 384 CTL_KERN, KERN_PROC_ARGS, CTL_EOL); 385 386 /* 387 "nodes" under these: 388 389 KERN_PROC_ALL 390 KERN_PROC_PID pid 391 KERN_PROC_PGRP pgrp 392 KERN_PROC_SESSION sess 393 KERN_PROC_TTY tty 394 KERN_PROC_UID uid 395 KERN_PROC_RUID uid 396 KERN_PROC_GID gid 397 KERN_PROC_RGID gid 398 399 all in all, probably not worth the effort... 400 */ 401 } 402 403 /* 404 * Initialize process 0. 405 */ 406 void 407 proc0_init(void) 408 { 409 struct proc *p; 410 struct pgrp *pg; 411 struct rlimit *rlim; 412 rlim_t lim; 413 int i; 414 415 p = &proc0; 416 pg = &pgrp0; 417 418 mutex_init(&p->p_stmutex, MUTEX_DEFAULT, IPL_HIGH); 419 mutex_init(&p->p_auxlock, MUTEX_DEFAULT, IPL_NONE); 420 p->p_lock = mutex_obj_alloc(MUTEX_DEFAULT, IPL_NONE); 421 422 rw_init(&p->p_reflock); 423 cv_init(&p->p_waitcv, "wait"); 424 cv_init(&p->p_lwpcv, "lwpwait"); 425 426 LIST_INSERT_HEAD(&p->p_lwps, &lwp0, l_sibling); 427 428 pid_table[0].pt_proc = p; 429 LIST_INSERT_HEAD(&allproc, p, p_list); 430 431 pid_table[0].pt_pgrp = pg; 432 LIST_INSERT_HEAD(&pg->pg_members, p, p_pglist); 433 434 #ifdef __HAVE_SYSCALL_INTERN 435 (*p->p_emul->e_syscall_intern)(p); 436 #endif 437 438 /* Create credentials. */ 439 cred0 = kauth_cred_alloc(); 440 p->p_cred = cred0; 441 442 /* Create the CWD info. */ 443 rw_init(&cwdi0.cwdi_lock); 444 445 /* Create the limits structures. */ 446 mutex_init(&limit0.pl_lock, MUTEX_DEFAULT, IPL_NONE); 447 448 rlim = limit0.pl_rlimit; 449 for (i = 0; i < __arraycount(limit0.pl_rlimit); i++) { 450 rlim[i].rlim_cur = RLIM_INFINITY; 451 rlim[i].rlim_max = RLIM_INFINITY; 452 } 453 454 rlim[RLIMIT_NOFILE].rlim_max = maxfiles; 455 rlim[RLIMIT_NOFILE].rlim_cur = maxfiles < nofile ? maxfiles : nofile; 456 457 rlim[RLIMIT_NPROC].rlim_max = maxproc; 458 rlim[RLIMIT_NPROC].rlim_cur = maxproc < maxuprc ? maxproc : maxuprc; 459 460 lim = MIN(VM_MAXUSER_ADDRESS, ctob((rlim_t)uvmexp.free)); 461 rlim[RLIMIT_RSS].rlim_max = lim; 462 rlim[RLIMIT_MEMLOCK].rlim_max = lim; 463 rlim[RLIMIT_MEMLOCK].rlim_cur = lim / 3; 464 465 /* Note that default core name has zero length. */ 466 limit0.pl_corename = defcorename; 467 limit0.pl_cnlen = 0; 468 limit0.pl_refcnt = 1; 469 limit0.pl_writeable = false; 470 limit0.pl_sv_limit = NULL; 471 472 /* Configure virtual memory system, set vm rlimits. */ 473 uvm_init_limits(p); 474 475 /* Initialize file descriptor table for proc0. */ 476 fd_init(&filedesc0); 477 478 /* 479 * Initialize proc0's vmspace, which uses the kernel pmap. 480 * All kernel processes (which never have user space mappings) 481 * share proc0's vmspace, and thus, the kernel pmap. 482 */ 483 uvmspace_init(&vmspace0, pmap_kernel(), round_page(VM_MIN_ADDRESS), 484 trunc_page(VM_MAX_ADDRESS)); 485 486 /* Initialize signal state for proc0. XXX IPL_SCHED */ 487 mutex_init(&p->p_sigacts->sa_mutex, MUTEX_DEFAULT, IPL_SCHED); 488 siginit(p); 489 490 proc_initspecific(p); 491 kdtrace_proc_ctor(NULL, p); 492 } 493 494 /* 495 * Session reference counting. 496 */ 497 498 void 499 proc_sesshold(struct session *ss) 500 { 501 502 KASSERT(mutex_owned(proc_lock)); 503 ss->s_count++; 504 } 505 506 void 507 proc_sessrele(struct session *ss) 508 { 509 510 KASSERT(mutex_owned(proc_lock)); 511 /* 512 * We keep the pgrp with the same id as the session in order to 513 * stop a process being given the same pid. Since the pgrp holds 514 * a reference to the session, it must be a 'zombie' pgrp by now. 515 */ 516 if (--ss->s_count == 0) { 517 struct pgrp *pg; 518 519 pg = pg_remove(ss->s_sid); 520 mutex_exit(proc_lock); 521 522 kmem_free(pg, sizeof(struct pgrp)); 523 kmem_free(ss, sizeof(struct session)); 524 } else { 525 mutex_exit(proc_lock); 526 } 527 } 528 529 /* 530 * Check that the specified process group is in the session of the 531 * specified process. 532 * Treats -ve ids as process ids. 533 * Used to validate TIOCSPGRP requests. 534 */ 535 int 536 pgid_in_session(struct proc *p, pid_t pg_id) 537 { 538 struct pgrp *pgrp; 539 struct session *session; 540 int error; 541 542 mutex_enter(proc_lock); 543 if (pg_id < 0) { 544 struct proc *p1 = proc_find(-pg_id); 545 if (p1 == NULL) { 546 error = EINVAL; 547 goto fail; 548 } 549 pgrp = p1->p_pgrp; 550 } else { 551 pgrp = pgrp_find(pg_id); 552 if (pgrp == NULL) { 553 error = EINVAL; 554 goto fail; 555 } 556 } 557 session = pgrp->pg_session; 558 error = (session != p->p_pgrp->pg_session) ? EPERM : 0; 559 fail: 560 mutex_exit(proc_lock); 561 return error; 562 } 563 564 /* 565 * p_inferior: is p an inferior of q? 566 */ 567 static inline bool 568 p_inferior(struct proc *p, struct proc *q) 569 { 570 571 KASSERT(mutex_owned(proc_lock)); 572 573 for (; p != q; p = p->p_pptr) 574 if (p->p_pid == 0) 575 return false; 576 return true; 577 } 578 579 /* 580 * proc_find: locate a process by the ID. 581 * 582 * => Must be called with proc_lock held. 583 */ 584 proc_t * 585 proc_find_raw(pid_t pid) 586 { 587 struct pid_table *pt; 588 proc_t *p; 589 590 KASSERT(mutex_owned(proc_lock)); 591 pt = &pid_table[pid & pid_tbl_mask]; 592 p = pt->pt_proc; 593 if (__predict_false(!P_VALID(p) || pt->pt_pid != pid)) { 594 return NULL; 595 } 596 return p; 597 } 598 599 proc_t * 600 proc_find(pid_t pid) 601 { 602 proc_t *p; 603 604 p = proc_find_raw(pid); 605 if (__predict_false(p == NULL)) { 606 return NULL; 607 } 608 609 /* 610 * Only allow live processes to be found by PID. 611 * XXX: p_stat might change, since unlocked. 612 */ 613 if (__predict_true(p->p_stat == SACTIVE || p->p_stat == SSTOP)) { 614 return p; 615 } 616 return NULL; 617 } 618 619 /* 620 * pgrp_find: locate a process group by the ID. 621 * 622 * => Must be called with proc_lock held. 623 */ 624 struct pgrp * 625 pgrp_find(pid_t pgid) 626 { 627 struct pgrp *pg; 628 629 KASSERT(mutex_owned(proc_lock)); 630 631 pg = pid_table[pgid & pid_tbl_mask].pt_pgrp; 632 633 /* 634 * Cannot look up a process group that only exists because the 635 * session has not died yet (traditional). 636 */ 637 if (pg == NULL || pg->pg_id != pgid || LIST_EMPTY(&pg->pg_members)) { 638 return NULL; 639 } 640 return pg; 641 } 642 643 static void 644 expand_pid_table(void) 645 { 646 size_t pt_size, tsz; 647 struct pid_table *n_pt, *new_pt; 648 struct proc *proc; 649 struct pgrp *pgrp; 650 pid_t pid, rpid; 651 u_int i; 652 uint new_pt_mask; 653 654 pt_size = pid_tbl_mask + 1; 655 tsz = pt_size * 2 * sizeof(struct pid_table); 656 new_pt = kmem_alloc(tsz, KM_SLEEP); 657 new_pt_mask = pt_size * 2 - 1; 658 659 mutex_enter(proc_lock); 660 if (pt_size != pid_tbl_mask + 1) { 661 /* Another process beat us to it... */ 662 mutex_exit(proc_lock); 663 kmem_free(new_pt, tsz); 664 return; 665 } 666 667 /* 668 * Copy entries from old table into new one. 669 * If 'pid' is 'odd' we need to place in the upper half, 670 * even pid's to the lower half. 671 * Free items stay in the low half so we don't have to 672 * fixup the reference to them. 673 * We stuff free items on the front of the freelist 674 * because we can't write to unmodified entries. 675 * Processing the table backwards maintains a semblance 676 * of issuing pid numbers that increase with time. 677 */ 678 i = pt_size - 1; 679 n_pt = new_pt + i; 680 for (; ; i--, n_pt--) { 681 proc = pid_table[i].pt_proc; 682 pgrp = pid_table[i].pt_pgrp; 683 if (!P_VALID(proc)) { 684 /* Up 'use count' so that link is valid */ 685 pid = (P_NEXT(proc) + pt_size) & ~pt_size; 686 rpid = 0; 687 proc = P_FREE(pid); 688 if (pgrp) 689 pid = pgrp->pg_id; 690 } else { 691 pid = pid_table[i].pt_pid; 692 rpid = pid; 693 } 694 695 /* Save entry in appropriate half of table */ 696 n_pt[pid & pt_size].pt_proc = proc; 697 n_pt[pid & pt_size].pt_pgrp = pgrp; 698 n_pt[pid & pt_size].pt_pid = rpid; 699 700 /* Put other piece on start of free list */ 701 pid = (pid ^ pt_size) & ~pid_tbl_mask; 702 n_pt[pid & pt_size].pt_proc = 703 P_FREE((pid & ~pt_size) | next_free_pt); 704 n_pt[pid & pt_size].pt_pgrp = 0; 705 n_pt[pid & pt_size].pt_pid = 0; 706 707 next_free_pt = i | (pid & pt_size); 708 if (i == 0) 709 break; 710 } 711 712 /* Save old table size and switch tables */ 713 tsz = pt_size * sizeof(struct pid_table); 714 n_pt = pid_table; 715 pid_table = new_pt; 716 pid_tbl_mask = new_pt_mask; 717 718 /* 719 * pid_max starts as PID_MAX (= 30000), once we have 16384 720 * allocated pids we need it to be larger! 721 */ 722 if (pid_tbl_mask > PID_MAX) { 723 pid_max = pid_tbl_mask * 2 + 1; 724 pid_alloc_lim |= pid_alloc_lim << 1; 725 } else 726 pid_alloc_lim <<= 1; /* doubles number of free slots... */ 727 728 mutex_exit(proc_lock); 729 kmem_free(n_pt, tsz); 730 } 731 732 struct proc * 733 proc_alloc(void) 734 { 735 struct proc *p; 736 737 p = pool_cache_get(proc_cache, PR_WAITOK); 738 p->p_stat = SIDL; /* protect against others */ 739 proc_initspecific(p); 740 kdtrace_proc_ctor(NULL, p); 741 p->p_pid = -1; 742 proc_alloc_pid(p); 743 return p; 744 } 745 746 pid_t 747 proc_alloc_pid(struct proc *p) 748 { 749 struct pid_table *pt; 750 pid_t pid; 751 int nxt; 752 753 for (;;expand_pid_table()) { 754 if (__predict_false(pid_alloc_cnt >= pid_alloc_lim)) 755 /* ensure pids cycle through 2000+ values */ 756 continue; 757 mutex_enter(proc_lock); 758 pt = &pid_table[next_free_pt]; 759 #ifdef DIAGNOSTIC 760 if (__predict_false(P_VALID(pt->pt_proc) || pt->pt_pgrp)) 761 panic("proc_alloc: slot busy"); 762 #endif 763 nxt = P_NEXT(pt->pt_proc); 764 if (nxt & pid_tbl_mask) 765 break; 766 /* Table full - expand (NB last entry not used....) */ 767 mutex_exit(proc_lock); 768 } 769 770 /* pid is 'saved use count' + 'size' + entry */ 771 pid = (nxt & ~pid_tbl_mask) + pid_tbl_mask + 1 + next_free_pt; 772 if ((uint)pid > (uint)pid_max) 773 pid &= pid_tbl_mask; 774 next_free_pt = nxt & pid_tbl_mask; 775 776 /* Grab table slot */ 777 pt->pt_proc = p; 778 779 KASSERT(pt->pt_pid == 0); 780 pt->pt_pid = pid; 781 if (p->p_pid == -1) { 782 p->p_pid = pid; 783 } 784 pid_alloc_cnt++; 785 mutex_exit(proc_lock); 786 787 return pid; 788 } 789 790 /* 791 * Free a process id - called from proc_free (in kern_exit.c) 792 * 793 * Called with the proc_lock held. 794 */ 795 void 796 proc_free_pid(pid_t pid) 797 { 798 struct pid_table *pt; 799 800 KASSERT(mutex_owned(proc_lock)); 801 802 pt = &pid_table[pid & pid_tbl_mask]; 803 804 /* save pid use count in slot */ 805 pt->pt_proc = P_FREE(pid & ~pid_tbl_mask); 806 KASSERT(pt->pt_pid == pid); 807 pt->pt_pid = 0; 808 809 if (pt->pt_pgrp == NULL) { 810 /* link last freed entry onto ours */ 811 pid &= pid_tbl_mask; 812 pt = &pid_table[last_free_pt]; 813 pt->pt_proc = P_FREE(P_NEXT(pt->pt_proc) | pid); 814 pt->pt_pid = 0; 815 last_free_pt = pid; 816 pid_alloc_cnt--; 817 } 818 819 atomic_dec_uint(&nprocs); 820 } 821 822 void 823 proc_free_mem(struct proc *p) 824 { 825 826 kdtrace_proc_dtor(NULL, p); 827 pool_cache_put(proc_cache, p); 828 } 829 830 /* 831 * proc_enterpgrp: move p to a new or existing process group (and session). 832 * 833 * If we are creating a new pgrp, the pgid should equal 834 * the calling process' pid. 835 * If is only valid to enter a process group that is in the session 836 * of the process. 837 * Also mksess should only be set if we are creating a process group 838 * 839 * Only called from sys_setsid and sys_setpgid. 840 */ 841 int 842 proc_enterpgrp(struct proc *curp, pid_t pid, pid_t pgid, bool mksess) 843 { 844 struct pgrp *new_pgrp, *pgrp; 845 struct session *sess; 846 struct proc *p; 847 int rval; 848 pid_t pg_id = NO_PGID; 849 850 sess = mksess ? kmem_alloc(sizeof(*sess), KM_SLEEP) : NULL; 851 852 /* Allocate data areas we might need before doing any validity checks */ 853 mutex_enter(proc_lock); /* Because pid_table might change */ 854 if (pid_table[pgid & pid_tbl_mask].pt_pgrp == 0) { 855 mutex_exit(proc_lock); 856 new_pgrp = kmem_alloc(sizeof(*new_pgrp), KM_SLEEP); 857 mutex_enter(proc_lock); 858 } else 859 new_pgrp = NULL; 860 rval = EPERM; /* most common error (to save typing) */ 861 862 /* Check pgrp exists or can be created */ 863 pgrp = pid_table[pgid & pid_tbl_mask].pt_pgrp; 864 if (pgrp != NULL && pgrp->pg_id != pgid) 865 goto done; 866 867 /* Can only set another process under restricted circumstances. */ 868 if (pid != curp->p_pid) { 869 /* Must exist and be one of our children... */ 870 p = proc_find(pid); 871 if (p == NULL || !p_inferior(p, curp)) { 872 rval = ESRCH; 873 goto done; 874 } 875 /* ... in the same session... */ 876 if (sess != NULL || p->p_session != curp->p_session) 877 goto done; 878 /* ... existing pgid must be in same session ... */ 879 if (pgrp != NULL && pgrp->pg_session != p->p_session) 880 goto done; 881 /* ... and not done an exec. */ 882 if (p->p_flag & PK_EXEC) { 883 rval = EACCES; 884 goto done; 885 } 886 } else { 887 /* ... setsid() cannot re-enter a pgrp */ 888 if (mksess && (curp->p_pgid == curp->p_pid || 889 pgrp_find(curp->p_pid))) 890 goto done; 891 p = curp; 892 } 893 894 /* Changing the process group/session of a session 895 leader is definitely off limits. */ 896 if (SESS_LEADER(p)) { 897 if (sess == NULL && p->p_pgrp == pgrp) 898 /* unless it's a definite noop */ 899 rval = 0; 900 goto done; 901 } 902 903 /* Can only create a process group with id of process */ 904 if (pgrp == NULL && pgid != pid) 905 goto done; 906 907 /* Can only create a session if creating pgrp */ 908 if (sess != NULL && pgrp != NULL) 909 goto done; 910 911 /* Check we allocated memory for a pgrp... */ 912 if (pgrp == NULL && new_pgrp == NULL) 913 goto done; 914 915 /* Don't attach to 'zombie' pgrp */ 916 if (pgrp != NULL && LIST_EMPTY(&pgrp->pg_members)) 917 goto done; 918 919 /* Expect to succeed now */ 920 rval = 0; 921 922 if (pgrp == p->p_pgrp) 923 /* nothing to do */ 924 goto done; 925 926 /* Ok all setup, link up required structures */ 927 928 if (pgrp == NULL) { 929 pgrp = new_pgrp; 930 new_pgrp = NULL; 931 if (sess != NULL) { 932 sess->s_sid = p->p_pid; 933 sess->s_leader = p; 934 sess->s_count = 1; 935 sess->s_ttyvp = NULL; 936 sess->s_ttyp = NULL; 937 sess->s_flags = p->p_session->s_flags & ~S_LOGIN_SET; 938 memcpy(sess->s_login, p->p_session->s_login, 939 sizeof(sess->s_login)); 940 p->p_lflag &= ~PL_CONTROLT; 941 } else { 942 sess = p->p_pgrp->pg_session; 943 proc_sesshold(sess); 944 } 945 pgrp->pg_session = sess; 946 sess = NULL; 947 948 pgrp->pg_id = pgid; 949 LIST_INIT(&pgrp->pg_members); 950 #ifdef DIAGNOSTIC 951 if (__predict_false(pid_table[pgid & pid_tbl_mask].pt_pgrp)) 952 panic("enterpgrp: pgrp table slot in use"); 953 if (__predict_false(mksess && p != curp)) 954 panic("enterpgrp: mksession and p != curproc"); 955 #endif 956 pid_table[pgid & pid_tbl_mask].pt_pgrp = pgrp; 957 pgrp->pg_jobc = 0; 958 } 959 960 /* 961 * Adjust eligibility of affected pgrps to participate in job control. 962 * Increment eligibility counts before decrementing, otherwise we 963 * could reach 0 spuriously during the first call. 964 */ 965 fixjobc(p, pgrp, 1); 966 fixjobc(p, p->p_pgrp, 0); 967 968 /* Interlock with ttread(). */ 969 mutex_spin_enter(&tty_lock); 970 971 /* Move process to requested group. */ 972 LIST_REMOVE(p, p_pglist); 973 if (LIST_EMPTY(&p->p_pgrp->pg_members)) 974 /* defer delete until we've dumped the lock */ 975 pg_id = p->p_pgrp->pg_id; 976 p->p_pgrp = pgrp; 977 LIST_INSERT_HEAD(&pgrp->pg_members, p, p_pglist); 978 979 /* Done with the swap; we can release the tty mutex. */ 980 mutex_spin_exit(&tty_lock); 981 982 done: 983 if (pg_id != NO_PGID) { 984 /* Releases proc_lock. */ 985 pg_delete(pg_id); 986 } else { 987 mutex_exit(proc_lock); 988 } 989 if (sess != NULL) 990 kmem_free(sess, sizeof(*sess)); 991 if (new_pgrp != NULL) 992 kmem_free(new_pgrp, sizeof(*new_pgrp)); 993 #ifdef DEBUG_PGRP 994 if (__predict_false(rval)) 995 printf("enterpgrp(%d,%d,%d), curproc %d, rval %d\n", 996 pid, pgid, mksess, curp->p_pid, rval); 997 #endif 998 return rval; 999 } 1000 1001 /* 1002 * proc_leavepgrp: remove a process from its process group. 1003 * => must be called with the proc_lock held, which will be released; 1004 */ 1005 void 1006 proc_leavepgrp(struct proc *p) 1007 { 1008 struct pgrp *pgrp; 1009 1010 KASSERT(mutex_owned(proc_lock)); 1011 1012 /* Interlock with ttread() */ 1013 mutex_spin_enter(&tty_lock); 1014 pgrp = p->p_pgrp; 1015 LIST_REMOVE(p, p_pglist); 1016 p->p_pgrp = NULL; 1017 mutex_spin_exit(&tty_lock); 1018 1019 if (LIST_EMPTY(&pgrp->pg_members)) { 1020 /* Releases proc_lock. */ 1021 pg_delete(pgrp->pg_id); 1022 } else { 1023 mutex_exit(proc_lock); 1024 } 1025 } 1026 1027 /* 1028 * pg_remove: remove a process group from the table. 1029 * => must be called with the proc_lock held; 1030 * => returns process group to free; 1031 */ 1032 static struct pgrp * 1033 pg_remove(pid_t pg_id) 1034 { 1035 struct pgrp *pgrp; 1036 struct pid_table *pt; 1037 1038 KASSERT(mutex_owned(proc_lock)); 1039 1040 pt = &pid_table[pg_id & pid_tbl_mask]; 1041 pgrp = pt->pt_pgrp; 1042 1043 KASSERT(pgrp != NULL); 1044 KASSERT(pgrp->pg_id == pg_id); 1045 KASSERT(LIST_EMPTY(&pgrp->pg_members)); 1046 1047 pt->pt_pgrp = NULL; 1048 1049 if (!P_VALID(pt->pt_proc)) { 1050 /* Orphaned pgrp, put slot onto free list. */ 1051 KASSERT((P_NEXT(pt->pt_proc) & pid_tbl_mask) == 0); 1052 pg_id &= pid_tbl_mask; 1053 pt = &pid_table[last_free_pt]; 1054 pt->pt_proc = P_FREE(P_NEXT(pt->pt_proc) | pg_id); 1055 KASSERT(pt->pt_pid == 0); 1056 last_free_pt = pg_id; 1057 pid_alloc_cnt--; 1058 } 1059 return pgrp; 1060 } 1061 1062 /* 1063 * pg_delete: delete and free a process group. 1064 * => must be called with the proc_lock held, which will be released. 1065 */ 1066 static void 1067 pg_delete(pid_t pg_id) 1068 { 1069 struct pgrp *pg; 1070 struct tty *ttyp; 1071 struct session *ss; 1072 1073 KASSERT(mutex_owned(proc_lock)); 1074 1075 pg = pid_table[pg_id & pid_tbl_mask].pt_pgrp; 1076 if (pg == NULL || pg->pg_id != pg_id || !LIST_EMPTY(&pg->pg_members)) { 1077 mutex_exit(proc_lock); 1078 return; 1079 } 1080 1081 ss = pg->pg_session; 1082 1083 /* Remove reference (if any) from tty to this process group */ 1084 mutex_spin_enter(&tty_lock); 1085 ttyp = ss->s_ttyp; 1086 if (ttyp != NULL && ttyp->t_pgrp == pg) { 1087 ttyp->t_pgrp = NULL; 1088 KASSERT(ttyp->t_session == ss); 1089 } 1090 mutex_spin_exit(&tty_lock); 1091 1092 /* 1093 * The leading process group in a session is freed by proc_sessrele(), 1094 * if last reference. Note: proc_sessrele() releases proc_lock. 1095 */ 1096 pg = (ss->s_sid != pg->pg_id) ? pg_remove(pg_id) : NULL; 1097 proc_sessrele(ss); 1098 1099 if (pg != NULL) { 1100 /* Free it, if was not done by proc_sessrele(). */ 1101 kmem_free(pg, sizeof(struct pgrp)); 1102 } 1103 } 1104 1105 /* 1106 * Adjust pgrp jobc counters when specified process changes process group. 1107 * We count the number of processes in each process group that "qualify" 1108 * the group for terminal job control (those with a parent in a different 1109 * process group of the same session). If that count reaches zero, the 1110 * process group becomes orphaned. Check both the specified process' 1111 * process group and that of its children. 1112 * entering == 0 => p is leaving specified group. 1113 * entering == 1 => p is entering specified group. 1114 * 1115 * Call with proc_lock held. 1116 */ 1117 void 1118 fixjobc(struct proc *p, struct pgrp *pgrp, int entering) 1119 { 1120 struct pgrp *hispgrp; 1121 struct session *mysession = pgrp->pg_session; 1122 struct proc *child; 1123 1124 KASSERT(mutex_owned(proc_lock)); 1125 1126 /* 1127 * Check p's parent to see whether p qualifies its own process 1128 * group; if so, adjust count for p's process group. 1129 */ 1130 hispgrp = p->p_pptr->p_pgrp; 1131 if (hispgrp != pgrp && hispgrp->pg_session == mysession) { 1132 if (entering) { 1133 pgrp->pg_jobc++; 1134 p->p_lflag &= ~PL_ORPHANPG; 1135 } else if (--pgrp->pg_jobc == 0) 1136 orphanpg(pgrp); 1137 } 1138 1139 /* 1140 * Check this process' children to see whether they qualify 1141 * their process groups; if so, adjust counts for children's 1142 * process groups. 1143 */ 1144 LIST_FOREACH(child, &p->p_children, p_sibling) { 1145 hispgrp = child->p_pgrp; 1146 if (hispgrp != pgrp && hispgrp->pg_session == mysession && 1147 !P_ZOMBIE(child)) { 1148 if (entering) { 1149 child->p_lflag &= ~PL_ORPHANPG; 1150 hispgrp->pg_jobc++; 1151 } else if (--hispgrp->pg_jobc == 0) 1152 orphanpg(hispgrp); 1153 } 1154 } 1155 } 1156 1157 /* 1158 * A process group has become orphaned; 1159 * if there are any stopped processes in the group, 1160 * hang-up all process in that group. 1161 * 1162 * Call with proc_lock held. 1163 */ 1164 static void 1165 orphanpg(struct pgrp *pg) 1166 { 1167 struct proc *p; 1168 1169 KASSERT(mutex_owned(proc_lock)); 1170 1171 LIST_FOREACH(p, &pg->pg_members, p_pglist) { 1172 if (p->p_stat == SSTOP) { 1173 p->p_lflag |= PL_ORPHANPG; 1174 psignal(p, SIGHUP); 1175 psignal(p, SIGCONT); 1176 } 1177 } 1178 } 1179 1180 #ifdef DDB 1181 #include <ddb/db_output.h> 1182 void pidtbl_dump(void); 1183 void 1184 pidtbl_dump(void) 1185 { 1186 struct pid_table *pt; 1187 struct proc *p; 1188 struct pgrp *pgrp; 1189 int id; 1190 1191 db_printf("pid table %p size %x, next %x, last %x\n", 1192 pid_table, pid_tbl_mask+1, 1193 next_free_pt, last_free_pt); 1194 for (pt = pid_table, id = 0; id <= pid_tbl_mask; id++, pt++) { 1195 p = pt->pt_proc; 1196 if (!P_VALID(p) && !pt->pt_pgrp) 1197 continue; 1198 db_printf(" id %x: ", id); 1199 if (P_VALID(p)) 1200 db_printf("slotpid %d proc %p id %d (0x%x) %s\n", 1201 pt->pt_pid, p, p->p_pid, p->p_pid, p->p_comm); 1202 else 1203 db_printf("next %x use %x\n", 1204 P_NEXT(p) & pid_tbl_mask, 1205 P_NEXT(p) & ~pid_tbl_mask); 1206 if ((pgrp = pt->pt_pgrp)) { 1207 db_printf("\tsession %p, sid %d, count %d, login %s\n", 1208 pgrp->pg_session, pgrp->pg_session->s_sid, 1209 pgrp->pg_session->s_count, 1210 pgrp->pg_session->s_login); 1211 db_printf("\tpgrp %p, pg_id %d, pg_jobc %d, members %p\n", 1212 pgrp, pgrp->pg_id, pgrp->pg_jobc, 1213 LIST_FIRST(&pgrp->pg_members)); 1214 LIST_FOREACH(p, &pgrp->pg_members, p_pglist) { 1215 db_printf("\t\tpid %d addr %p pgrp %p %s\n", 1216 p->p_pid, p, p->p_pgrp, p->p_comm); 1217 } 1218 } 1219 } 1220 } 1221 #endif /* DDB */ 1222 1223 #ifdef KSTACK_CHECK_MAGIC 1224 1225 #define KSTACK_MAGIC 0xdeadbeaf 1226 1227 /* XXX should be per process basis? */ 1228 static int kstackleftmin = KSTACK_SIZE; 1229 static int kstackleftthres = KSTACK_SIZE / 8; 1230 1231 void 1232 kstack_setup_magic(const struct lwp *l) 1233 { 1234 uint32_t *ip; 1235 uint32_t const *end; 1236 1237 KASSERT(l != NULL); 1238 KASSERT(l != &lwp0); 1239 1240 /* 1241 * fill all the stack with magic number 1242 * so that later modification on it can be detected. 1243 */ 1244 ip = (uint32_t *)KSTACK_LOWEST_ADDR(l); 1245 end = (uint32_t *)((char *)KSTACK_LOWEST_ADDR(l) + KSTACK_SIZE); 1246 for (; ip < end; ip++) { 1247 *ip = KSTACK_MAGIC; 1248 } 1249 } 1250 1251 void 1252 kstack_check_magic(const struct lwp *l) 1253 { 1254 uint32_t const *ip, *end; 1255 int stackleft; 1256 1257 KASSERT(l != NULL); 1258 1259 /* don't check proc0 */ /*XXX*/ 1260 if (l == &lwp0) 1261 return; 1262 1263 #ifdef __MACHINE_STACK_GROWS_UP 1264 /* stack grows upwards (eg. hppa) */ 1265 ip = (uint32_t *)((void *)KSTACK_LOWEST_ADDR(l) + KSTACK_SIZE); 1266 end = (uint32_t *)KSTACK_LOWEST_ADDR(l); 1267 for (ip--; ip >= end; ip--) 1268 if (*ip != KSTACK_MAGIC) 1269 break; 1270 1271 stackleft = (void *)KSTACK_LOWEST_ADDR(l) + KSTACK_SIZE - (void *)ip; 1272 #else /* __MACHINE_STACK_GROWS_UP */ 1273 /* stack grows downwards (eg. i386) */ 1274 ip = (uint32_t *)KSTACK_LOWEST_ADDR(l); 1275 end = (uint32_t *)((char *)KSTACK_LOWEST_ADDR(l) + KSTACK_SIZE); 1276 for (; ip < end; ip++) 1277 if (*ip != KSTACK_MAGIC) 1278 break; 1279 1280 stackleft = ((const char *)ip) - (const char *)KSTACK_LOWEST_ADDR(l); 1281 #endif /* __MACHINE_STACK_GROWS_UP */ 1282 1283 if (kstackleftmin > stackleft) { 1284 kstackleftmin = stackleft; 1285 if (stackleft < kstackleftthres) 1286 printf("warning: kernel stack left %d bytes" 1287 "(pid %u:lid %u)\n", stackleft, 1288 (u_int)l->l_proc->p_pid, (u_int)l->l_lid); 1289 } 1290 1291 if (stackleft <= 0) { 1292 panic("magic on the top of kernel stack changed for " 1293 "pid %u, lid %u: maybe kernel stack overflow", 1294 (u_int)l->l_proc->p_pid, (u_int)l->l_lid); 1295 } 1296 } 1297 #endif /* KSTACK_CHECK_MAGIC */ 1298 1299 int 1300 proclist_foreach_call(struct proclist *list, 1301 int (*callback)(struct proc *, void *arg), void *arg) 1302 { 1303 struct proc marker; 1304 struct proc *p; 1305 int ret = 0; 1306 1307 marker.p_flag = PK_MARKER; 1308 mutex_enter(proc_lock); 1309 for (p = LIST_FIRST(list); ret == 0 && p != NULL;) { 1310 if (p->p_flag & PK_MARKER) { 1311 p = LIST_NEXT(p, p_list); 1312 continue; 1313 } 1314 LIST_INSERT_AFTER(p, &marker, p_list); 1315 ret = (*callback)(p, arg); 1316 KASSERT(mutex_owned(proc_lock)); 1317 p = LIST_NEXT(&marker, p_list); 1318 LIST_REMOVE(&marker, p_list); 1319 } 1320 mutex_exit(proc_lock); 1321 1322 return ret; 1323 } 1324 1325 int 1326 proc_vmspace_getref(struct proc *p, struct vmspace **vm) 1327 { 1328 1329 /* XXXCDC: how should locking work here? */ 1330 1331 /* curproc exception is for coredump. */ 1332 1333 if ((p != curproc && (p->p_sflag & PS_WEXIT) != 0) || 1334 (p->p_vmspace->vm_refcnt < 1)) { /* XXX */ 1335 return EFAULT; 1336 } 1337 1338 uvmspace_addref(p->p_vmspace); 1339 *vm = p->p_vmspace; 1340 1341 return 0; 1342 } 1343 1344 /* 1345 * Acquire a write lock on the process credential. 1346 */ 1347 void 1348 proc_crmod_enter(void) 1349 { 1350 struct lwp *l = curlwp; 1351 struct proc *p = l->l_proc; 1352 kauth_cred_t oc; 1353 1354 /* Reset what needs to be reset in plimit. */ 1355 if (p->p_limit->pl_corename != defcorename) { 1356 lim_setcorename(p, defcorename, 0); 1357 } 1358 1359 mutex_enter(p->p_lock); 1360 1361 /* Ensure the LWP cached credentials are up to date. */ 1362 if ((oc = l->l_cred) != p->p_cred) { 1363 kauth_cred_hold(p->p_cred); 1364 l->l_cred = p->p_cred; 1365 kauth_cred_free(oc); 1366 } 1367 } 1368 1369 /* 1370 * Set in a new process credential, and drop the write lock. The credential 1371 * must have a reference already. Optionally, free a no-longer required 1372 * credential. The scheduler also needs to inspect p_cred, so we also 1373 * briefly acquire the sched state mutex. 1374 */ 1375 void 1376 proc_crmod_leave(kauth_cred_t scred, kauth_cred_t fcred, bool sugid) 1377 { 1378 struct lwp *l = curlwp, *l2; 1379 struct proc *p = l->l_proc; 1380 kauth_cred_t oc; 1381 1382 KASSERT(mutex_owned(p->p_lock)); 1383 1384 /* Is there a new credential to set in? */ 1385 if (scred != NULL) { 1386 p->p_cred = scred; 1387 LIST_FOREACH(l2, &p->p_lwps, l_sibling) { 1388 if (l2 != l) 1389 l2->l_prflag |= LPR_CRMOD; 1390 } 1391 1392 /* Ensure the LWP cached credentials are up to date. */ 1393 if ((oc = l->l_cred) != scred) { 1394 kauth_cred_hold(scred); 1395 l->l_cred = scred; 1396 } 1397 } else 1398 oc = NULL; /* XXXgcc */ 1399 1400 if (sugid) { 1401 /* 1402 * Mark process as having changed credentials, stops 1403 * tracing etc. 1404 */ 1405 p->p_flag |= PK_SUGID; 1406 } 1407 1408 mutex_exit(p->p_lock); 1409 1410 /* If there is a credential to be released, free it now. */ 1411 if (fcred != NULL) { 1412 KASSERT(scred != NULL); 1413 kauth_cred_free(fcred); 1414 if (oc != scred) 1415 kauth_cred_free(oc); 1416 } 1417 } 1418 1419 /* 1420 * proc_specific_key_create -- 1421 * Create a key for subsystem proc-specific data. 1422 */ 1423 int 1424 proc_specific_key_create(specificdata_key_t *keyp, specificdata_dtor_t dtor) 1425 { 1426 1427 return (specificdata_key_create(proc_specificdata_domain, keyp, dtor)); 1428 } 1429 1430 /* 1431 * proc_specific_key_delete -- 1432 * Delete a key for subsystem proc-specific data. 1433 */ 1434 void 1435 proc_specific_key_delete(specificdata_key_t key) 1436 { 1437 1438 specificdata_key_delete(proc_specificdata_domain, key); 1439 } 1440 1441 /* 1442 * proc_initspecific -- 1443 * Initialize a proc's specificdata container. 1444 */ 1445 void 1446 proc_initspecific(struct proc *p) 1447 { 1448 int error; 1449 1450 error = specificdata_init(proc_specificdata_domain, &p->p_specdataref); 1451 KASSERT(error == 0); 1452 } 1453 1454 /* 1455 * proc_finispecific -- 1456 * Finalize a proc's specificdata container. 1457 */ 1458 void 1459 proc_finispecific(struct proc *p) 1460 { 1461 1462 specificdata_fini(proc_specificdata_domain, &p->p_specdataref); 1463 } 1464 1465 /* 1466 * proc_getspecific -- 1467 * Return proc-specific data corresponding to the specified key. 1468 */ 1469 void * 1470 proc_getspecific(struct proc *p, specificdata_key_t key) 1471 { 1472 1473 return (specificdata_getspecific(proc_specificdata_domain, 1474 &p->p_specdataref, key)); 1475 } 1476 1477 /* 1478 * proc_setspecific -- 1479 * Set proc-specific data corresponding to the specified key. 1480 */ 1481 void 1482 proc_setspecific(struct proc *p, specificdata_key_t key, void *data) 1483 { 1484 1485 specificdata_setspecific(proc_specificdata_domain, 1486 &p->p_specdataref, key, data); 1487 } 1488 1489 int 1490 proc_uidmatch(kauth_cred_t cred, kauth_cred_t target) 1491 { 1492 int r = 0; 1493 1494 if (kauth_cred_getuid(cred) != kauth_cred_getuid(target) || 1495 kauth_cred_getuid(cred) != kauth_cred_getsvuid(target)) { 1496 /* 1497 * suid proc of ours or proc not ours 1498 */ 1499 r = EPERM; 1500 } else if (kauth_cred_getgid(target) != kauth_cred_getsvgid(target)) { 1501 /* 1502 * sgid proc has sgid back to us temporarily 1503 */ 1504 r = EPERM; 1505 } else { 1506 /* 1507 * our rgid must be in target's group list (ie, 1508 * sub-processes started by a sgid process) 1509 */ 1510 int ismember = 0; 1511 1512 if (kauth_cred_ismember_gid(cred, 1513 kauth_cred_getgid(target), &ismember) != 0 || 1514 !ismember) 1515 r = EPERM; 1516 } 1517 1518 return (r); 1519 } 1520 1521 /* 1522 * sysctl stuff 1523 */ 1524 1525 #define KERN_PROCSLOP (5 * sizeof(struct kinfo_proc)) 1526 1527 static const u_int sysctl_flagmap[] = { 1528 PK_ADVLOCK, P_ADVLOCK, 1529 PK_EXEC, P_EXEC, 1530 PK_NOCLDWAIT, P_NOCLDWAIT, 1531 PK_32, P_32, 1532 PK_CLDSIGIGN, P_CLDSIGIGN, 1533 PK_SUGID, P_SUGID, 1534 0 1535 }; 1536 1537 static const u_int sysctl_sflagmap[] = { 1538 PS_NOCLDSTOP, P_NOCLDSTOP, 1539 PS_WEXIT, P_WEXIT, 1540 PS_STOPFORK, P_STOPFORK, 1541 PS_STOPEXEC, P_STOPEXEC, 1542 PS_STOPEXIT, P_STOPEXIT, 1543 0 1544 }; 1545 1546 static const u_int sysctl_slflagmap[] = { 1547 PSL_TRACED, P_TRACED, 1548 PSL_FSTRACE, P_FSTRACE, 1549 PSL_CHTRACED, P_CHTRACED, 1550 PSL_SYSCALL, P_SYSCALL, 1551 0 1552 }; 1553 1554 static const u_int sysctl_lflagmap[] = { 1555 PL_CONTROLT, P_CONTROLT, 1556 PL_PPWAIT, P_PPWAIT, 1557 0 1558 }; 1559 1560 static const u_int sysctl_stflagmap[] = { 1561 PST_PROFIL, P_PROFIL, 1562 0 1563 1564 }; 1565 1566 /* used by kern_lwp also */ 1567 const u_int sysctl_lwpflagmap[] = { 1568 LW_SINTR, L_SINTR, 1569 LW_SYSTEM, L_SYSTEM, 1570 LW_SA, L_SA, /* WRS ??? */ 1571 0 1572 }; 1573 1574 /* 1575 * Find the most ``active'' lwp of a process and return it for ps display 1576 * purposes 1577 */ 1578 static struct lwp * 1579 proc_active_lwp(struct proc *p) 1580 { 1581 static const int ostat[] = { 1582 0, 1583 2, /* LSIDL */ 1584 6, /* LSRUN */ 1585 5, /* LSSLEEP */ 1586 4, /* LSSTOP */ 1587 0, /* LSZOMB */ 1588 1, /* LSDEAD */ 1589 7, /* LSONPROC */ 1590 3 /* LSSUSPENDED */ 1591 }; 1592 1593 struct lwp *l, *lp = NULL; 1594 LIST_FOREACH(l, &p->p_lwps, l_sibling) { 1595 KASSERT(l->l_stat >= 0 && l->l_stat < __arraycount(ostat)); 1596 if (lp == NULL || 1597 ostat[l->l_stat] > ostat[lp->l_stat] || 1598 (ostat[l->l_stat] == ostat[lp->l_stat] && 1599 l->l_cpticks > lp->l_cpticks)) { 1600 lp = l; 1601 continue; 1602 } 1603 } 1604 return lp; 1605 } 1606 1607 static int 1608 sysctl_doeproc(SYSCTLFN_ARGS) 1609 { 1610 union { 1611 struct kinfo_proc kproc; 1612 struct kinfo_proc2 kproc2; 1613 } *kbuf; 1614 struct proc *p, *next, *marker; 1615 char *where, *dp; 1616 int type, op, arg, error; 1617 u_int elem_size, kelem_size, elem_count; 1618 size_t buflen, needed; 1619 bool match, zombie, mmmbrains; 1620 1621 if (namelen == 1 && name[0] == CTL_QUERY) 1622 return (sysctl_query(SYSCTLFN_CALL(rnode))); 1623 1624 dp = where = oldp; 1625 buflen = where != NULL ? *oldlenp : 0; 1626 error = 0; 1627 needed = 0; 1628 type = rnode->sysctl_num; 1629 1630 if (type == KERN_PROC) { 1631 if (namelen != 2 && !(namelen == 1 && name[0] == KERN_PROC_ALL)) 1632 return (EINVAL); 1633 op = name[0]; 1634 if (op != KERN_PROC_ALL) 1635 arg = name[1]; 1636 else 1637 arg = 0; /* Quell compiler warning */ 1638 elem_count = 0; /* Ditto */ 1639 kelem_size = elem_size = sizeof(kbuf->kproc); 1640 } else { 1641 if (namelen != 4) 1642 return (EINVAL); 1643 op = name[0]; 1644 arg = name[1]; 1645 elem_size = name[2]; 1646 elem_count = name[3]; 1647 kelem_size = sizeof(kbuf->kproc2); 1648 } 1649 1650 sysctl_unlock(); 1651 1652 kbuf = kmem_alloc(sizeof(*kbuf), KM_SLEEP); 1653 marker = kmem_alloc(sizeof(*marker), KM_SLEEP); 1654 marker->p_flag = PK_MARKER; 1655 1656 mutex_enter(proc_lock); 1657 mmmbrains = false; 1658 for (p = LIST_FIRST(&allproc);; p = next) { 1659 if (p == NULL) { 1660 if (!mmmbrains) { 1661 p = LIST_FIRST(&zombproc); 1662 mmmbrains = true; 1663 } 1664 if (p == NULL) 1665 break; 1666 } 1667 next = LIST_NEXT(p, p_list); 1668 if ((p->p_flag & PK_MARKER) != 0) 1669 continue; 1670 1671 /* 1672 * Skip embryonic processes. 1673 */ 1674 if (p->p_stat == SIDL) 1675 continue; 1676 1677 mutex_enter(p->p_lock); 1678 error = kauth_authorize_process(l->l_cred, 1679 KAUTH_PROCESS_CANSEE, p, 1680 KAUTH_ARG(KAUTH_REQ_PROCESS_CANSEE_ENTRY), NULL, NULL); 1681 if (error != 0) { 1682 mutex_exit(p->p_lock); 1683 continue; 1684 } 1685 1686 /* 1687 * TODO - make more efficient (see notes below). 1688 * do by session. 1689 */ 1690 switch (op) { 1691 case KERN_PROC_PID: 1692 /* could do this with just a lookup */ 1693 match = (p->p_pid == (pid_t)arg); 1694 break; 1695 1696 case KERN_PROC_PGRP: 1697 /* could do this by traversing pgrp */ 1698 match = (p->p_pgrp->pg_id == (pid_t)arg); 1699 break; 1700 1701 case KERN_PROC_SESSION: 1702 match = (p->p_session->s_sid == (pid_t)arg); 1703 break; 1704 1705 case KERN_PROC_TTY: 1706 match = true; 1707 if (arg == (int) KERN_PROC_TTY_REVOKE) { 1708 if ((p->p_lflag & PL_CONTROLT) == 0 || 1709 p->p_session->s_ttyp == NULL || 1710 p->p_session->s_ttyvp != NULL) { 1711 match = false; 1712 } 1713 } else if ((p->p_lflag & PL_CONTROLT) == 0 || 1714 p->p_session->s_ttyp == NULL) { 1715 if ((dev_t)arg != KERN_PROC_TTY_NODEV) { 1716 match = false; 1717 } 1718 } else if (p->p_session->s_ttyp->t_dev != (dev_t)arg) { 1719 match = false; 1720 } 1721 break; 1722 1723 case KERN_PROC_UID: 1724 match = (kauth_cred_geteuid(p->p_cred) == (uid_t)arg); 1725 break; 1726 1727 case KERN_PROC_RUID: 1728 match = (kauth_cred_getuid(p->p_cred) == (uid_t)arg); 1729 break; 1730 1731 case KERN_PROC_GID: 1732 match = (kauth_cred_getegid(p->p_cred) == (uid_t)arg); 1733 break; 1734 1735 case KERN_PROC_RGID: 1736 match = (kauth_cred_getgid(p->p_cred) == (uid_t)arg); 1737 break; 1738 1739 case KERN_PROC_ALL: 1740 match = true; 1741 /* allow everything */ 1742 break; 1743 1744 default: 1745 error = EINVAL; 1746 mutex_exit(p->p_lock); 1747 goto cleanup; 1748 } 1749 if (!match) { 1750 mutex_exit(p->p_lock); 1751 continue; 1752 } 1753 1754 /* 1755 * Grab a hold on the process. 1756 */ 1757 if (mmmbrains) { 1758 zombie = true; 1759 } else { 1760 zombie = !rw_tryenter(&p->p_reflock, RW_READER); 1761 } 1762 if (zombie) { 1763 LIST_INSERT_AFTER(p, marker, p_list); 1764 } 1765 1766 if (buflen >= elem_size && 1767 (type == KERN_PROC || elem_count > 0)) { 1768 if (type == KERN_PROC) { 1769 kbuf->kproc.kp_proc = *p; 1770 fill_eproc(p, &kbuf->kproc.kp_eproc, zombie); 1771 } else { 1772 fill_kproc2(p, &kbuf->kproc2, zombie); 1773 elem_count--; 1774 } 1775 mutex_exit(p->p_lock); 1776 mutex_exit(proc_lock); 1777 /* 1778 * Copy out elem_size, but not larger than kelem_size 1779 */ 1780 error = sysctl_copyout(l, kbuf, dp, 1781 min(kelem_size, elem_size)); 1782 mutex_enter(proc_lock); 1783 if (error) { 1784 goto bah; 1785 } 1786 dp += elem_size; 1787 buflen -= elem_size; 1788 } else { 1789 mutex_exit(p->p_lock); 1790 } 1791 needed += elem_size; 1792 1793 /* 1794 * Release reference to process. 1795 */ 1796 if (zombie) { 1797 next = LIST_NEXT(marker, p_list); 1798 LIST_REMOVE(marker, p_list); 1799 } else { 1800 rw_exit(&p->p_reflock); 1801 next = LIST_NEXT(p, p_list); 1802 } 1803 } 1804 mutex_exit(proc_lock); 1805 1806 if (where != NULL) { 1807 *oldlenp = dp - where; 1808 if (needed > *oldlenp) { 1809 error = ENOMEM; 1810 goto out; 1811 } 1812 } else { 1813 needed += KERN_PROCSLOP; 1814 *oldlenp = needed; 1815 } 1816 if (kbuf) 1817 kmem_free(kbuf, sizeof(*kbuf)); 1818 if (marker) 1819 kmem_free(marker, sizeof(*marker)); 1820 sysctl_relock(); 1821 return 0; 1822 bah: 1823 if (zombie) 1824 LIST_REMOVE(marker, p_list); 1825 else 1826 rw_exit(&p->p_reflock); 1827 cleanup: 1828 mutex_exit(proc_lock); 1829 out: 1830 if (kbuf) 1831 kmem_free(kbuf, sizeof(*kbuf)); 1832 if (marker) 1833 kmem_free(marker, sizeof(*marker)); 1834 sysctl_relock(); 1835 return error; 1836 } 1837 1838 int 1839 copyin_psstrings(struct proc *p, struct ps_strings *arginfo) 1840 { 1841 1842 #ifdef COMPAT_NETBSD32 1843 if (p->p_flag & PK_32) { 1844 struct ps_strings32 arginfo32; 1845 1846 int error = copyin_proc(p, (void *)p->p_psstrp, &arginfo32, 1847 sizeof(arginfo32)); 1848 if (error) 1849 return error; 1850 arginfo->ps_argvstr = (void *)(uintptr_t)arginfo32.ps_argvstr; 1851 arginfo->ps_nargvstr = arginfo32.ps_nargvstr; 1852 arginfo->ps_envstr = (void *)(uintptr_t)arginfo32.ps_envstr; 1853 arginfo->ps_nenvstr = arginfo32.ps_nenvstr; 1854 return 0; 1855 } 1856 #endif 1857 return copyin_proc(p, (void *)p->p_psstrp, arginfo, sizeof(*arginfo)); 1858 } 1859 1860 static int 1861 copy_procargs_sysctl_cb(void *cookie_, const void *src, size_t off, size_t len) 1862 { 1863 void **cookie = cookie_; 1864 struct lwp *l = cookie[0]; 1865 char *dst = cookie[1]; 1866 1867 return sysctl_copyout(l, src, dst + off, len); 1868 } 1869 1870 /* 1871 * sysctl helper routine for kern.proc_args pseudo-subtree. 1872 */ 1873 static int 1874 sysctl_kern_proc_args(SYSCTLFN_ARGS) 1875 { 1876 struct ps_strings pss; 1877 struct proc *p; 1878 pid_t pid; 1879 int type, error; 1880 void *cookie[2]; 1881 1882 if (namelen == 1 && name[0] == CTL_QUERY) 1883 return (sysctl_query(SYSCTLFN_CALL(rnode))); 1884 1885 if (newp != NULL || namelen != 2) 1886 return (EINVAL); 1887 pid = name[0]; 1888 type = name[1]; 1889 1890 switch (type) { 1891 case KERN_PROC_ARGV: 1892 case KERN_PROC_NARGV: 1893 case KERN_PROC_ENV: 1894 case KERN_PROC_NENV: 1895 /* ok */ 1896 break; 1897 default: 1898 return (EINVAL); 1899 } 1900 1901 sysctl_unlock(); 1902 1903 /* check pid */ 1904 mutex_enter(proc_lock); 1905 if ((p = proc_find(pid)) == NULL) { 1906 error = EINVAL; 1907 goto out_locked; 1908 } 1909 mutex_enter(p->p_lock); 1910 1911 /* Check permission. */ 1912 if (type == KERN_PROC_ARGV || type == KERN_PROC_NARGV) 1913 error = kauth_authorize_process(l->l_cred, KAUTH_PROCESS_CANSEE, 1914 p, KAUTH_ARG(KAUTH_REQ_PROCESS_CANSEE_ARGS), NULL, NULL); 1915 else if (type == KERN_PROC_ENV || type == KERN_PROC_NENV) 1916 error = kauth_authorize_process(l->l_cred, KAUTH_PROCESS_CANSEE, 1917 p, KAUTH_ARG(KAUTH_REQ_PROCESS_CANSEE_ENV), NULL, NULL); 1918 else 1919 error = EINVAL; /* XXXGCC */ 1920 if (error) { 1921 mutex_exit(p->p_lock); 1922 goto out_locked; 1923 } 1924 1925 if (oldp == NULL) { 1926 if (type == KERN_PROC_NARGV || type == KERN_PROC_NENV) 1927 *oldlenp = sizeof (int); 1928 else 1929 *oldlenp = ARG_MAX; /* XXX XXX XXX */ 1930 error = 0; 1931 mutex_exit(p->p_lock); 1932 goto out_locked; 1933 } 1934 1935 /* 1936 * Zombies don't have a stack, so we can't read their psstrings. 1937 * System processes also don't have a user stack. 1938 */ 1939 if (P_ZOMBIE(p) || (p->p_flag & PK_SYSTEM) != 0) { 1940 error = EINVAL; 1941 mutex_exit(p->p_lock); 1942 goto out_locked; 1943 } 1944 1945 error = rw_tryenter(&p->p_reflock, RW_READER) ? 0 : EBUSY; 1946 mutex_exit(p->p_lock); 1947 if (error) { 1948 goto out_locked; 1949 } 1950 mutex_exit(proc_lock); 1951 1952 if (type == KERN_PROC_NARGV || type == KERN_PROC_NENV) { 1953 int value; 1954 if ((error = copyin_psstrings(p, &pss)) == 0) { 1955 if (type == KERN_PROC_NARGV) 1956 value = pss.ps_nargvstr; 1957 else 1958 value = pss.ps_nenvstr; 1959 error = sysctl_copyout(l, &value, oldp, sizeof(value)); 1960 *oldlenp = sizeof(value); 1961 } 1962 } else { 1963 cookie[0] = l; 1964 cookie[1] = oldp; 1965 error = copy_procargs(p, type, oldlenp, 1966 copy_procargs_sysctl_cb, cookie); 1967 } 1968 rw_exit(&p->p_reflock); 1969 sysctl_relock(); 1970 return error; 1971 1972 out_locked: 1973 mutex_exit(proc_lock); 1974 sysctl_relock(); 1975 return error; 1976 } 1977 1978 int 1979 copy_procargs(struct proc *p, int oid, size_t *limit, 1980 int (*cb)(void *, const void *, size_t, size_t), void *cookie) 1981 { 1982 struct ps_strings pss; 1983 size_t len, i, loaded, entry_len; 1984 struct uio auio; 1985 struct iovec aiov; 1986 int error, argvlen; 1987 char *arg; 1988 char **argv; 1989 vaddr_t user_argv; 1990 struct vmspace *vmspace; 1991 1992 /* 1993 * Allocate a temporary buffer to hold the argument vector and 1994 * the arguments themselve. 1995 */ 1996 arg = kmem_alloc(PAGE_SIZE, KM_SLEEP); 1997 argv = kmem_alloc(PAGE_SIZE, KM_SLEEP); 1998 1999 /* 2000 * Lock the process down in memory. 2001 */ 2002 vmspace = p->p_vmspace; 2003 uvmspace_addref(vmspace); 2004 2005 /* 2006 * Read in the ps_strings structure. 2007 */ 2008 if ((error = copyin_psstrings(p, &pss)) != 0) 2009 goto done; 2010 2011 /* 2012 * Now read the address of the argument vector. 2013 */ 2014 switch (oid) { 2015 case KERN_PROC_ARGV: 2016 user_argv = (uintptr_t)pss.ps_argvstr; 2017 argvlen = pss.ps_nargvstr; 2018 break; 2019 case KERN_PROC_ENV: 2020 user_argv = (uintptr_t)pss.ps_envstr; 2021 argvlen = pss.ps_nenvstr; 2022 break; 2023 default: 2024 error = EINVAL; 2025 goto done; 2026 } 2027 2028 if (argvlen < 0) { 2029 error = EIO; 2030 goto done; 2031 } 2032 2033 #ifdef COMPAT_NETBSD32 2034 if (p->p_flag & PK_32) 2035 entry_len = sizeof(netbsd32_charp); 2036 else 2037 #endif 2038 entry_len = sizeof(char *); 2039 2040 /* 2041 * Now copy each string. 2042 */ 2043 len = 0; /* bytes written to user buffer */ 2044 loaded = 0; /* bytes from argv already processed */ 2045 i = 0; /* To make compiler happy */ 2046 2047 for (; argvlen; --argvlen) { 2048 int finished = 0; 2049 vaddr_t base; 2050 size_t xlen; 2051 int j; 2052 2053 if (loaded == 0) { 2054 size_t rem = entry_len * argvlen; 2055 loaded = MIN(rem, PAGE_SIZE); 2056 error = copyin_vmspace(vmspace, 2057 (const void *)user_argv, argv, loaded); 2058 if (error) 2059 break; 2060 user_argv += loaded; 2061 i = 0; 2062 } 2063 2064 #ifdef COMPAT_NETBSD32 2065 if (p->p_flag & PK_32) { 2066 netbsd32_charp *argv32; 2067 2068 argv32 = (netbsd32_charp *)argv; 2069 base = (vaddr_t)NETBSD32PTR64(argv32[i++]); 2070 } else 2071 #endif 2072 base = (vaddr_t)argv[i++]; 2073 loaded -= entry_len; 2074 2075 /* 2076 * The program has messed around with its arguments, 2077 * possibly deleting some, and replacing them with 2078 * NULL's. Treat this as the last argument and not 2079 * a failure. 2080 */ 2081 if (base == 0) 2082 break; 2083 2084 while (!finished) { 2085 xlen = PAGE_SIZE - (base & PAGE_MASK); 2086 2087 aiov.iov_base = arg; 2088 aiov.iov_len = PAGE_SIZE; 2089 auio.uio_iov = &aiov; 2090 auio.uio_iovcnt = 1; 2091 auio.uio_offset = base; 2092 auio.uio_resid = xlen; 2093 auio.uio_rw = UIO_READ; 2094 UIO_SETUP_SYSSPACE(&auio); 2095 error = uvm_io(&vmspace->vm_map, &auio); 2096 if (error) 2097 goto done; 2098 2099 /* Look for the end of the string */ 2100 for (j = 0; j < xlen; j++) { 2101 if (arg[j] == '\0') { 2102 xlen = j + 1; 2103 finished = 1; 2104 break; 2105 } 2106 } 2107 2108 /* Check for user buffer overflow */ 2109 if (len + xlen > *limit) { 2110 finished = 1; 2111 if (len > *limit) 2112 xlen = 0; 2113 else 2114 xlen = *limit - len; 2115 } 2116 2117 /* Copyout the page */ 2118 error = (*cb)(cookie, arg, len, xlen); 2119 if (error) 2120 goto done; 2121 2122 len += xlen; 2123 base += xlen; 2124 } 2125 } 2126 *limit = len; 2127 2128 done: 2129 kmem_free(argv, PAGE_SIZE); 2130 kmem_free(arg, PAGE_SIZE); 2131 uvmspace_free(vmspace); 2132 return error; 2133 } 2134 2135 /* 2136 * Fill in an eproc structure for the specified process. 2137 */ 2138 void 2139 fill_eproc(struct proc *p, struct eproc *ep, bool zombie) 2140 { 2141 struct tty *tp; 2142 struct lwp *l; 2143 2144 KASSERT(mutex_owned(proc_lock)); 2145 KASSERT(mutex_owned(p->p_lock)); 2146 2147 memset(ep, 0, sizeof(*ep)); 2148 2149 ep->e_paddr = p; 2150 ep->e_sess = p->p_session; 2151 if (p->p_cred) { 2152 kauth_cred_topcred(p->p_cred, &ep->e_pcred); 2153 kauth_cred_toucred(p->p_cred, &ep->e_ucred); 2154 } 2155 if (p->p_stat != SIDL && !P_ZOMBIE(p) && !zombie) { 2156 struct vmspace *vm = p->p_vmspace; 2157 2158 ep->e_vm.vm_rssize = vm_resident_count(vm); 2159 ep->e_vm.vm_tsize = vm->vm_tsize; 2160 ep->e_vm.vm_dsize = vm->vm_dsize; 2161 ep->e_vm.vm_ssize = vm->vm_ssize; 2162 ep->e_vm.vm_map.size = vm->vm_map.size; 2163 2164 /* Pick the primary (first) LWP */ 2165 l = proc_active_lwp(p); 2166 KASSERT(l != NULL); 2167 lwp_lock(l); 2168 if (l->l_wchan) 2169 strncpy(ep->e_wmesg, l->l_wmesg, WMESGLEN); 2170 lwp_unlock(l); 2171 } 2172 if (p->p_pptr) 2173 ep->e_ppid = p->p_pptr->p_pid; 2174 if (p->p_pgrp && p->p_session) { 2175 ep->e_pgid = p->p_pgrp->pg_id; 2176 ep->e_jobc = p->p_pgrp->pg_jobc; 2177 ep->e_sid = p->p_session->s_sid; 2178 if ((p->p_lflag & PL_CONTROLT) && 2179 (tp = ep->e_sess->s_ttyp)) { 2180 ep->e_tdev = tp->t_dev; 2181 ep->e_tpgid = tp->t_pgrp ? tp->t_pgrp->pg_id : NO_PGID; 2182 ep->e_tsess = tp->t_session; 2183 } else 2184 ep->e_tdev = (uint32_t)NODEV; 2185 ep->e_flag = ep->e_sess->s_ttyvp ? EPROC_CTTY : 0; 2186 if (SESS_LEADER(p)) 2187 ep->e_flag |= EPROC_SLEADER; 2188 strncpy(ep->e_login, ep->e_sess->s_login, MAXLOGNAME); 2189 } 2190 ep->e_xsize = ep->e_xrssize = 0; 2191 ep->e_xccount = ep->e_xswrss = 0; 2192 } 2193 2194 /* 2195 * Fill in a kinfo_proc2 structure for the specified process. 2196 */ 2197 static void 2198 fill_kproc2(struct proc *p, struct kinfo_proc2 *ki, bool zombie) 2199 { 2200 struct tty *tp; 2201 struct lwp *l, *l2; 2202 struct timeval ut, st, rt; 2203 sigset_t ss1, ss2; 2204 struct rusage ru; 2205 struct vmspace *vm; 2206 2207 KASSERT(mutex_owned(proc_lock)); 2208 KASSERT(mutex_owned(p->p_lock)); 2209 2210 sigemptyset(&ss1); 2211 sigemptyset(&ss2); 2212 memset(ki, 0, sizeof(*ki)); 2213 2214 ki->p_paddr = PTRTOUINT64(p); 2215 ki->p_fd = PTRTOUINT64(p->p_fd); 2216 ki->p_cwdi = PTRTOUINT64(p->p_cwdi); 2217 ki->p_stats = PTRTOUINT64(p->p_stats); 2218 ki->p_limit = PTRTOUINT64(p->p_limit); 2219 ki->p_vmspace = PTRTOUINT64(p->p_vmspace); 2220 ki->p_sigacts = PTRTOUINT64(p->p_sigacts); 2221 ki->p_sess = PTRTOUINT64(p->p_session); 2222 ki->p_tsess = 0; /* may be changed if controlling tty below */ 2223 ki->p_ru = PTRTOUINT64(&p->p_stats->p_ru); 2224 ki->p_eflag = 0; 2225 ki->p_exitsig = p->p_exitsig; 2226 ki->p_flag = L_INMEM; /* Process never swapped out */ 2227 ki->p_flag |= sysctl_map_flags(sysctl_flagmap, p->p_flag); 2228 ki->p_flag |= sysctl_map_flags(sysctl_sflagmap, p->p_sflag); 2229 ki->p_flag |= sysctl_map_flags(sysctl_slflagmap, p->p_slflag); 2230 ki->p_flag |= sysctl_map_flags(sysctl_lflagmap, p->p_lflag); 2231 ki->p_flag |= sysctl_map_flags(sysctl_stflagmap, p->p_stflag); 2232 ki->p_pid = p->p_pid; 2233 if (p->p_pptr) 2234 ki->p_ppid = p->p_pptr->p_pid; 2235 else 2236 ki->p_ppid = 0; 2237 ki->p_uid = kauth_cred_geteuid(p->p_cred); 2238 ki->p_ruid = kauth_cred_getuid(p->p_cred); 2239 ki->p_gid = kauth_cred_getegid(p->p_cred); 2240 ki->p_rgid = kauth_cred_getgid(p->p_cred); 2241 ki->p_svuid = kauth_cred_getsvuid(p->p_cred); 2242 ki->p_svgid = kauth_cred_getsvgid(p->p_cred); 2243 ki->p_ngroups = kauth_cred_ngroups(p->p_cred); 2244 kauth_cred_getgroups(p->p_cred, ki->p_groups, 2245 min(ki->p_ngroups, sizeof(ki->p_groups) / sizeof(ki->p_groups[0])), 2246 UIO_SYSSPACE); 2247 2248 ki->p_uticks = p->p_uticks; 2249 ki->p_sticks = p->p_sticks; 2250 ki->p_iticks = p->p_iticks; 2251 ki->p_tpgid = NO_PGID; /* may be changed if controlling tty below */ 2252 ki->p_tracep = PTRTOUINT64(p->p_tracep); 2253 ki->p_traceflag = p->p_traceflag; 2254 2255 memcpy(&ki->p_sigignore, &p->p_sigctx.ps_sigignore,sizeof(ki_sigset_t)); 2256 memcpy(&ki->p_sigcatch, &p->p_sigctx.ps_sigcatch, sizeof(ki_sigset_t)); 2257 2258 ki->p_cpticks = 0; 2259 ki->p_pctcpu = p->p_pctcpu; 2260 ki->p_estcpu = 0; 2261 ki->p_stat = p->p_stat; /* Will likely be overridden by LWP status */ 2262 ki->p_realstat = p->p_stat; 2263 ki->p_nice = p->p_nice; 2264 ki->p_xstat = p->p_xstat; 2265 ki->p_acflag = p->p_acflag; 2266 2267 strncpy(ki->p_comm, p->p_comm, 2268 min(sizeof(ki->p_comm), sizeof(p->p_comm))); 2269 strncpy(ki->p_ename, p->p_emul->e_name, sizeof(ki->p_ename)); 2270 2271 ki->p_nlwps = p->p_nlwps; 2272 ki->p_realflag = ki->p_flag; 2273 2274 if (p->p_stat != SIDL && !P_ZOMBIE(p) && !zombie) { 2275 vm = p->p_vmspace; 2276 ki->p_vm_rssize = vm_resident_count(vm); 2277 ki->p_vm_tsize = vm->vm_tsize; 2278 ki->p_vm_dsize = vm->vm_dsize; 2279 ki->p_vm_ssize = vm->vm_ssize; 2280 ki->p_vm_vsize = vm->vm_map.size; 2281 /* 2282 * Since the stack is initially mapped mostly with 2283 * PROT_NONE and grown as needed, adjust the "mapped size" 2284 * to skip the unused stack portion. 2285 */ 2286 ki->p_vm_msize = 2287 atop(vm->vm_map.size) - vm->vm_issize + vm->vm_ssize; 2288 2289 /* Pick the primary (first) LWP */ 2290 l = proc_active_lwp(p); 2291 KASSERT(l != NULL); 2292 lwp_lock(l); 2293 ki->p_nrlwps = p->p_nrlwps; 2294 ki->p_forw = 0; 2295 ki->p_back = 0; 2296 ki->p_addr = PTRTOUINT64(l->l_addr); 2297 ki->p_stat = l->l_stat; 2298 ki->p_flag |= sysctl_map_flags(sysctl_lwpflagmap, l->l_flag); 2299 ki->p_swtime = l->l_swtime; 2300 ki->p_slptime = l->l_slptime; 2301 if (l->l_stat == LSONPROC) 2302 ki->p_schedflags = l->l_cpu->ci_schedstate.spc_flags; 2303 else 2304 ki->p_schedflags = 0; 2305 ki->p_priority = lwp_eprio(l); 2306 ki->p_usrpri = l->l_priority; 2307 if (l->l_wchan) 2308 strncpy(ki->p_wmesg, l->l_wmesg, sizeof(ki->p_wmesg)); 2309 ki->p_wchan = PTRTOUINT64(l->l_wchan); 2310 ki->p_cpuid = cpu_index(l->l_cpu); 2311 lwp_unlock(l); 2312 LIST_FOREACH(l, &p->p_lwps, l_sibling) { 2313 /* This is hardly correct, but... */ 2314 sigplusset(&l->l_sigpend.sp_set, &ss1); 2315 sigplusset(&l->l_sigmask, &ss2); 2316 ki->p_cpticks += l->l_cpticks; 2317 ki->p_pctcpu += l->l_pctcpu; 2318 ki->p_estcpu += l->l_estcpu; 2319 } 2320 } 2321 sigplusset(&p->p_sigpend.sp_set, &ss2); 2322 memcpy(&ki->p_siglist, &ss1, sizeof(ki_sigset_t)); 2323 memcpy(&ki->p_sigmask, &ss2, sizeof(ki_sigset_t)); 2324 2325 if (p->p_session != NULL) { 2326 ki->p_sid = p->p_session->s_sid; 2327 ki->p__pgid = p->p_pgrp->pg_id; 2328 if (p->p_session->s_ttyvp) 2329 ki->p_eflag |= EPROC_CTTY; 2330 if (SESS_LEADER(p)) 2331 ki->p_eflag |= EPROC_SLEADER; 2332 strncpy(ki->p_login, p->p_session->s_login, 2333 min(sizeof ki->p_login - 1, sizeof p->p_session->s_login)); 2334 ki->p_jobc = p->p_pgrp->pg_jobc; 2335 if ((p->p_lflag & PL_CONTROLT) && (tp = p->p_session->s_ttyp)) { 2336 ki->p_tdev = tp->t_dev; 2337 ki->p_tpgid = tp->t_pgrp ? tp->t_pgrp->pg_id : NO_PGID; 2338 ki->p_tsess = PTRTOUINT64(tp->t_session); 2339 } else { 2340 ki->p_tdev = (int32_t)NODEV; 2341 } 2342 } 2343 2344 if (!P_ZOMBIE(p) && !zombie) { 2345 ki->p_uvalid = 1; 2346 ki->p_ustart_sec = p->p_stats->p_start.tv_sec; 2347 ki->p_ustart_usec = p->p_stats->p_start.tv_usec; 2348 2349 calcru(p, &ut, &st, NULL, &rt); 2350 ki->p_rtime_sec = rt.tv_sec; 2351 ki->p_rtime_usec = rt.tv_usec; 2352 ki->p_uutime_sec = ut.tv_sec; 2353 ki->p_uutime_usec = ut.tv_usec; 2354 ki->p_ustime_sec = st.tv_sec; 2355 ki->p_ustime_usec = st.tv_usec; 2356 2357 memcpy(&ru, &p->p_stats->p_ru, sizeof(ru)); 2358 ki->p_uru_nvcsw = 0; 2359 ki->p_uru_nivcsw = 0; 2360 LIST_FOREACH(l2, &p->p_lwps, l_sibling) { 2361 ki->p_uru_nvcsw += (l2->l_ncsw - l2->l_nivcsw); 2362 ki->p_uru_nivcsw += l2->l_nivcsw; 2363 ruadd(&ru, &l2->l_ru); 2364 } 2365 ki->p_uru_maxrss = ru.ru_maxrss; 2366 ki->p_uru_ixrss = ru.ru_ixrss; 2367 ki->p_uru_idrss = ru.ru_idrss; 2368 ki->p_uru_isrss = ru.ru_isrss; 2369 ki->p_uru_minflt = ru.ru_minflt; 2370 ki->p_uru_majflt = ru.ru_majflt; 2371 ki->p_uru_nswap = ru.ru_nswap; 2372 ki->p_uru_inblock = ru.ru_inblock; 2373 ki->p_uru_oublock = ru.ru_oublock; 2374 ki->p_uru_msgsnd = ru.ru_msgsnd; 2375 ki->p_uru_msgrcv = ru.ru_msgrcv; 2376 ki->p_uru_nsignals = ru.ru_nsignals; 2377 2378 timeradd(&p->p_stats->p_cru.ru_utime, 2379 &p->p_stats->p_cru.ru_stime, &ut); 2380 ki->p_uctime_sec = ut.tv_sec; 2381 ki->p_uctime_usec = ut.tv_usec; 2382 } 2383 } 2384