1 /*- 2 * SPDX-License-Identifier: BSD-2-Clause-FreeBSD 3 * 4 * Copyright (C) 2001 Julian Elischer <julian@freebsd.org>. 5 * All rights reserved. 6 * 7 * Redistribution and use in source and binary forms, with or without 8 * modification, are permitted provided that the following conditions 9 * are met: 10 * 1. Redistributions of source code must retain the above copyright 11 * notice(s), this list of conditions and the following disclaimer as 12 * the first lines of this file unmodified other than the possible 13 * addition of one or more copyright notices. 14 * 2. Redistributions in binary form must reproduce the above copyright 15 * notice(s), this list of conditions and the following disclaimer in the 16 * documentation and/or other materials provided with the distribution. 17 * 18 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDER(S) ``AS IS'' AND ANY 19 * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED 20 * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE 21 * DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER(S) BE LIABLE FOR ANY 22 * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES 23 * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR 24 * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER 25 * CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 26 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 27 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH 28 * DAMAGE. 29 */ 30 31 #include "opt_witness.h" 32 #include "opt_hwpmc_hooks.h" 33 34 #include <sys/cdefs.h> 35 __FBSDID("$FreeBSD$"); 36 37 #include <sys/param.h> 38 #include <sys/systm.h> 39 #include <sys/kernel.h> 40 #include <sys/lock.h> 41 #include <sys/mutex.h> 42 #include <sys/proc.h> 43 #include <sys/rangelock.h> 44 #include <sys/resourcevar.h> 45 #include <sys/sdt.h> 46 #include <sys/smp.h> 47 #include <sys/sched.h> 48 #include <sys/sleepqueue.h> 49 #include <sys/selinfo.h> 50 #include <sys/syscallsubr.h> 51 #include <sys/sysent.h> 52 #include <sys/turnstile.h> 53 #include <sys/ktr.h> 54 #include <sys/rwlock.h> 55 #include <sys/umtx.h> 56 #include <sys/vmmeter.h> 57 #include <sys/cpuset.h> 58 #ifdef HWPMC_HOOKS 59 #include <sys/pmckern.h> 60 #endif 61 62 #include <security/audit/audit.h> 63 64 #include <vm/vm.h> 65 #include <vm/vm_extern.h> 66 #include <vm/uma.h> 67 #include <sys/eventhandler.h> 68 69 /* 70 * Asserts below verify the stability of struct thread and struct proc 71 * layout, as exposed by KBI to modules. On head, the KBI is allowed 72 * to drift, change to the structures must be accompanied by the 73 * assert update. 74 * 75 * On the stable branches after KBI freeze, conditions must not be 76 * violated. Typically new fields are moved to the end of the 77 * structures. 78 */ 79 #ifdef __amd64__ 80 _Static_assert(offsetof(struct thread, td_flags) == 0xfc, 81 "struct thread KBI td_flags"); 82 _Static_assert(offsetof(struct thread, td_pflags) == 0x104, 83 "struct thread KBI td_pflags"); 84 _Static_assert(offsetof(struct thread, td_frame) == 0x470, 85 "struct thread KBI td_frame"); 86 _Static_assert(offsetof(struct thread, td_emuldata) == 0x518, 87 "struct thread KBI td_emuldata"); 88 _Static_assert(offsetof(struct proc, p_flag) == 0xb0, 89 "struct proc KBI p_flag"); 90 _Static_assert(offsetof(struct proc, p_pid) == 0xbc, 91 "struct proc KBI p_pid"); 92 _Static_assert(offsetof(struct proc, p_filemon) == 0x3d0, 93 "struct proc KBI p_filemon"); 94 _Static_assert(offsetof(struct proc, p_comm) == 0x3e4, 95 "struct proc KBI p_comm"); 96 _Static_assert(offsetof(struct proc, p_emuldata) == 0x4b8, 97 "struct proc KBI p_emuldata"); 98 #endif 99 #ifdef __i386__ 100 _Static_assert(offsetof(struct thread, td_flags) == 0x98, 101 "struct thread KBI td_flags"); 102 _Static_assert(offsetof(struct thread, td_pflags) == 0xa0, 103 "struct thread KBI td_pflags"); 104 _Static_assert(offsetof(struct thread, td_frame) == 0x2e8, 105 "struct thread KBI td_frame"); 106 _Static_assert(offsetof(struct thread, td_emuldata) == 0x334, 107 "struct thread KBI td_emuldata"); 108 _Static_assert(offsetof(struct proc, p_flag) == 0x68, 109 "struct proc KBI p_flag"); 110 _Static_assert(offsetof(struct proc, p_pid) == 0x74, 111 "struct proc KBI p_pid"); 112 _Static_assert(offsetof(struct proc, p_filemon) == 0x27c, 113 "struct proc KBI p_filemon"); 114 _Static_assert(offsetof(struct proc, p_comm) == 0x28c, 115 "struct proc KBI p_comm"); 116 _Static_assert(offsetof(struct proc, p_emuldata) == 0x318, 117 "struct proc KBI p_emuldata"); 118 #endif 119 120 SDT_PROVIDER_DECLARE(proc); 121 SDT_PROBE_DEFINE(proc, , , lwp__exit); 122 123 /* 124 * thread related storage. 125 */ 126 static uma_zone_t thread_zone; 127 128 TAILQ_HEAD(, thread) zombie_threads = TAILQ_HEAD_INITIALIZER(zombie_threads); 129 static struct mtx zombie_lock; 130 MTX_SYSINIT(zombie_lock, &zombie_lock, "zombie lock", MTX_SPIN); 131 132 static void thread_zombie(struct thread *); 133 static int thread_unsuspend_one(struct thread *td, struct proc *p, 134 bool boundary); 135 136 #define TID_BUFFER_SIZE 1024 137 138 struct mtx tid_lock; 139 static struct unrhdr *tid_unrhdr; 140 static lwpid_t tid_buffer[TID_BUFFER_SIZE]; 141 static int tid_head, tid_tail; 142 static MALLOC_DEFINE(M_TIDHASH, "tidhash", "thread hash"); 143 144 struct tidhashhead *tidhashtbl; 145 u_long tidhash; 146 struct rwlock tidhash_lock; 147 148 EVENTHANDLER_LIST_DEFINE(thread_ctor); 149 EVENTHANDLER_LIST_DEFINE(thread_dtor); 150 EVENTHANDLER_LIST_DEFINE(thread_init); 151 EVENTHANDLER_LIST_DEFINE(thread_fini); 152 153 static lwpid_t 154 tid_alloc(void) 155 { 156 lwpid_t tid; 157 158 tid = alloc_unr(tid_unrhdr); 159 if (tid != -1) 160 return (tid); 161 mtx_lock(&tid_lock); 162 if (tid_head == tid_tail) { 163 mtx_unlock(&tid_lock); 164 return (-1); 165 } 166 tid = tid_buffer[tid_head]; 167 tid_head = (tid_head + 1) % TID_BUFFER_SIZE; 168 mtx_unlock(&tid_lock); 169 return (tid); 170 } 171 172 static void 173 tid_free(lwpid_t tid) 174 { 175 lwpid_t tmp_tid = -1; 176 177 mtx_lock(&tid_lock); 178 if ((tid_tail + 1) % TID_BUFFER_SIZE == tid_head) { 179 tmp_tid = tid_buffer[tid_head]; 180 tid_head = (tid_head + 1) % TID_BUFFER_SIZE; 181 } 182 tid_buffer[tid_tail] = tid; 183 tid_tail = (tid_tail + 1) % TID_BUFFER_SIZE; 184 mtx_unlock(&tid_lock); 185 if (tmp_tid != -1) 186 free_unr(tid_unrhdr, tmp_tid); 187 } 188 189 /* 190 * Prepare a thread for use. 191 */ 192 static int 193 thread_ctor(void *mem, int size, void *arg, int flags) 194 { 195 struct thread *td; 196 197 td = (struct thread *)mem; 198 td->td_state = TDS_INACTIVE; 199 td->td_oncpu = NOCPU; 200 201 td->td_tid = tid_alloc(); 202 203 /* 204 * Note that td_critnest begins life as 1 because the thread is not 205 * running and is thereby implicitly waiting to be on the receiving 206 * end of a context switch. 207 */ 208 td->td_critnest = 1; 209 td->td_lend_user_pri = PRI_MAX; 210 EVENTHANDLER_DIRECT_INVOKE(thread_ctor, td); 211 #ifdef AUDIT 212 audit_thread_alloc(td); 213 #endif 214 umtx_thread_alloc(td); 215 return (0); 216 } 217 218 /* 219 * Reclaim a thread after use. 220 */ 221 static void 222 thread_dtor(void *mem, int size, void *arg) 223 { 224 struct thread *td; 225 226 td = (struct thread *)mem; 227 228 #ifdef INVARIANTS 229 /* Verify that this thread is in a safe state to free. */ 230 switch (td->td_state) { 231 case TDS_INHIBITED: 232 case TDS_RUNNING: 233 case TDS_CAN_RUN: 234 case TDS_RUNQ: 235 /* 236 * We must never unlink a thread that is in one of 237 * these states, because it is currently active. 238 */ 239 panic("bad state for thread unlinking"); 240 /* NOTREACHED */ 241 case TDS_INACTIVE: 242 break; 243 default: 244 panic("bad thread state"); 245 /* NOTREACHED */ 246 } 247 #endif 248 #ifdef AUDIT 249 audit_thread_free(td); 250 #endif 251 /* Free all OSD associated to this thread. */ 252 osd_thread_exit(td); 253 td_softdep_cleanup(td); 254 MPASS(td->td_su == NULL); 255 256 EVENTHANDLER_DIRECT_INVOKE(thread_dtor, td); 257 tid_free(td->td_tid); 258 } 259 260 /* 261 * Initialize type-stable parts of a thread (when newly created). 262 */ 263 static int 264 thread_init(void *mem, int size, int flags) 265 { 266 struct thread *td; 267 268 td = (struct thread *)mem; 269 270 td->td_sleepqueue = sleepq_alloc(); 271 td->td_turnstile = turnstile_alloc(); 272 td->td_rlqe = NULL; 273 EVENTHANDLER_DIRECT_INVOKE(thread_init, td); 274 umtx_thread_init(td); 275 td->td_kstack = 0; 276 td->td_sel = NULL; 277 return (0); 278 } 279 280 /* 281 * Tear down type-stable parts of a thread (just before being discarded). 282 */ 283 static void 284 thread_fini(void *mem, int size) 285 { 286 struct thread *td; 287 288 td = (struct thread *)mem; 289 EVENTHANDLER_DIRECT_INVOKE(thread_fini, td); 290 rlqentry_free(td->td_rlqe); 291 turnstile_free(td->td_turnstile); 292 sleepq_free(td->td_sleepqueue); 293 umtx_thread_fini(td); 294 seltdfini(td); 295 } 296 297 /* 298 * For a newly created process, 299 * link up all the structures and its initial threads etc. 300 * called from: 301 * {arch}/{arch}/machdep.c {arch}_init(), init386() etc. 302 * proc_dtor() (should go away) 303 * proc_init() 304 */ 305 void 306 proc_linkup0(struct proc *p, struct thread *td) 307 { 308 TAILQ_INIT(&p->p_threads); /* all threads in proc */ 309 proc_linkup(p, td); 310 } 311 312 void 313 proc_linkup(struct proc *p, struct thread *td) 314 { 315 316 sigqueue_init(&p->p_sigqueue, p); 317 p->p_ksi = ksiginfo_alloc(1); 318 if (p->p_ksi != NULL) { 319 /* XXX p_ksi may be null if ksiginfo zone is not ready */ 320 p->p_ksi->ksi_flags = KSI_EXT | KSI_INS; 321 } 322 LIST_INIT(&p->p_mqnotifier); 323 p->p_numthreads = 0; 324 thread_link(td, p); 325 } 326 327 /* 328 * Initialize global thread allocation resources. 329 */ 330 void 331 threadinit(void) 332 { 333 334 mtx_init(&tid_lock, "TID lock", NULL, MTX_DEF); 335 336 /* 337 * pid_max cannot be greater than PID_MAX. 338 * leave one number for thread0. 339 */ 340 tid_unrhdr = new_unrhdr(PID_MAX + 2, INT_MAX, &tid_lock); 341 342 thread_zone = uma_zcreate("THREAD", sched_sizeof_thread(), 343 thread_ctor, thread_dtor, thread_init, thread_fini, 344 32 - 1, UMA_ZONE_NOFREE); 345 tidhashtbl = hashinit(maxproc / 2, M_TIDHASH, &tidhash); 346 rw_init(&tidhash_lock, "tidhash"); 347 } 348 349 /* 350 * Place an unused thread on the zombie list. 351 * Use the slpq as that must be unused by now. 352 */ 353 void 354 thread_zombie(struct thread *td) 355 { 356 mtx_lock_spin(&zombie_lock); 357 TAILQ_INSERT_HEAD(&zombie_threads, td, td_slpq); 358 mtx_unlock_spin(&zombie_lock); 359 } 360 361 /* 362 * Release a thread that has exited after cpu_throw(). 363 */ 364 void 365 thread_stash(struct thread *td) 366 { 367 atomic_subtract_rel_int(&td->td_proc->p_exitthreads, 1); 368 thread_zombie(td); 369 } 370 371 /* 372 * Reap zombie resources. 373 */ 374 void 375 thread_reap(void) 376 { 377 struct thread *td_first, *td_next; 378 379 /* 380 * Don't even bother to lock if none at this instant, 381 * we really don't care about the next instant. 382 */ 383 if (!TAILQ_EMPTY(&zombie_threads)) { 384 mtx_lock_spin(&zombie_lock); 385 td_first = TAILQ_FIRST(&zombie_threads); 386 if (td_first) 387 TAILQ_INIT(&zombie_threads); 388 mtx_unlock_spin(&zombie_lock); 389 while (td_first) { 390 td_next = TAILQ_NEXT(td_first, td_slpq); 391 thread_cow_free(td_first); 392 thread_free(td_first); 393 td_first = td_next; 394 } 395 } 396 } 397 398 /* 399 * Allocate a thread. 400 */ 401 struct thread * 402 thread_alloc(int pages) 403 { 404 struct thread *td; 405 406 thread_reap(); /* check if any zombies to get */ 407 408 td = (struct thread *)uma_zalloc(thread_zone, M_WAITOK); 409 KASSERT(td->td_kstack == 0, ("thread_alloc got thread with kstack")); 410 if (!vm_thread_new(td, pages)) { 411 uma_zfree(thread_zone, td); 412 return (NULL); 413 } 414 cpu_thread_alloc(td); 415 return (td); 416 } 417 418 int 419 thread_alloc_stack(struct thread *td, int pages) 420 { 421 422 KASSERT(td->td_kstack == 0, 423 ("thread_alloc_stack called on a thread with kstack")); 424 if (!vm_thread_new(td, pages)) 425 return (0); 426 cpu_thread_alloc(td); 427 return (1); 428 } 429 430 /* 431 * Deallocate a thread. 432 */ 433 void 434 thread_free(struct thread *td) 435 { 436 437 lock_profile_thread_exit(td); 438 if (td->td_cpuset) 439 cpuset_rel(td->td_cpuset); 440 td->td_cpuset = NULL; 441 cpu_thread_free(td); 442 if (td->td_kstack != 0) 443 vm_thread_dispose(td); 444 callout_drain(&td->td_slpcallout); 445 uma_zfree(thread_zone, td); 446 } 447 448 void 449 thread_cow_get_proc(struct thread *newtd, struct proc *p) 450 { 451 452 PROC_LOCK_ASSERT(p, MA_OWNED); 453 newtd->td_ucred = crhold(p->p_ucred); 454 newtd->td_limit = lim_hold(p->p_limit); 455 newtd->td_cowgen = p->p_cowgen; 456 } 457 458 void 459 thread_cow_get(struct thread *newtd, struct thread *td) 460 { 461 462 newtd->td_ucred = crhold(td->td_ucred); 463 newtd->td_limit = lim_hold(td->td_limit); 464 newtd->td_cowgen = td->td_cowgen; 465 } 466 467 void 468 thread_cow_free(struct thread *td) 469 { 470 471 if (td->td_ucred != NULL) 472 crfree(td->td_ucred); 473 if (td->td_limit != NULL) 474 lim_free(td->td_limit); 475 } 476 477 void 478 thread_cow_update(struct thread *td) 479 { 480 struct proc *p; 481 struct ucred *oldcred; 482 struct plimit *oldlimit; 483 484 p = td->td_proc; 485 oldcred = NULL; 486 oldlimit = NULL; 487 PROC_LOCK(p); 488 if (td->td_ucred != p->p_ucred) { 489 oldcred = td->td_ucred; 490 td->td_ucred = crhold(p->p_ucred); 491 } 492 if (td->td_limit != p->p_limit) { 493 oldlimit = td->td_limit; 494 td->td_limit = lim_hold(p->p_limit); 495 } 496 td->td_cowgen = p->p_cowgen; 497 PROC_UNLOCK(p); 498 if (oldcred != NULL) 499 crfree(oldcred); 500 if (oldlimit != NULL) 501 lim_free(oldlimit); 502 } 503 504 /* 505 * Discard the current thread and exit from its context. 506 * Always called with scheduler locked. 507 * 508 * Because we can't free a thread while we're operating under its context, 509 * push the current thread into our CPU's deadthread holder. This means 510 * we needn't worry about someone else grabbing our context before we 511 * do a cpu_throw(). 512 */ 513 void 514 thread_exit(void) 515 { 516 uint64_t runtime, new_switchtime; 517 struct thread *td; 518 struct thread *td2; 519 struct proc *p; 520 int wakeup_swapper; 521 522 td = curthread; 523 p = td->td_proc; 524 525 PROC_SLOCK_ASSERT(p, MA_OWNED); 526 mtx_assert(&Giant, MA_NOTOWNED); 527 528 PROC_LOCK_ASSERT(p, MA_OWNED); 529 KASSERT(p != NULL, ("thread exiting without a process")); 530 CTR3(KTR_PROC, "thread_exit: thread %p (pid %ld, %s)", td, 531 (long)p->p_pid, td->td_name); 532 SDT_PROBE0(proc, , , lwp__exit); 533 KASSERT(TAILQ_EMPTY(&td->td_sigqueue.sq_list), ("signal pending")); 534 535 #ifdef AUDIT 536 AUDIT_SYSCALL_EXIT(0, td); 537 #endif 538 /* 539 * drop FPU & debug register state storage, or any other 540 * architecture specific resources that 541 * would not be on a new untouched process. 542 */ 543 cpu_thread_exit(td); 544 545 /* 546 * The last thread is left attached to the process 547 * So that the whole bundle gets recycled. Skip 548 * all this stuff if we never had threads. 549 * EXIT clears all sign of other threads when 550 * it goes to single threading, so the last thread always 551 * takes the short path. 552 */ 553 if (p->p_flag & P_HADTHREADS) { 554 if (p->p_numthreads > 1) { 555 atomic_add_int(&td->td_proc->p_exitthreads, 1); 556 thread_unlink(td); 557 td2 = FIRST_THREAD_IN_PROC(p); 558 sched_exit_thread(td2, td); 559 560 /* 561 * The test below is NOT true if we are the 562 * sole exiting thread. P_STOPPED_SINGLE is unset 563 * in exit1() after it is the only survivor. 564 */ 565 if (P_SHOULDSTOP(p) == P_STOPPED_SINGLE) { 566 if (p->p_numthreads == p->p_suspcount) { 567 thread_lock(p->p_singlethread); 568 wakeup_swapper = thread_unsuspend_one( 569 p->p_singlethread, p, false); 570 thread_unlock(p->p_singlethread); 571 if (wakeup_swapper) 572 kick_proc0(); 573 } 574 } 575 576 PCPU_SET(deadthread, td); 577 } else { 578 /* 579 * The last thread is exiting.. but not through exit() 580 */ 581 panic ("thread_exit: Last thread exiting on its own"); 582 } 583 } 584 #ifdef HWPMC_HOOKS 585 /* 586 * If this thread is part of a process that is being tracked by hwpmc(4), 587 * inform the module of the thread's impending exit. 588 */ 589 if (PMC_PROC_IS_USING_PMCS(td->td_proc)) { 590 PMC_SWITCH_CONTEXT(td, PMC_FN_CSW_OUT); 591 PMC_CALL_HOOK_UNLOCKED(td, PMC_FN_THR_EXIT, NULL); 592 } 593 #endif 594 PROC_UNLOCK(p); 595 PROC_STATLOCK(p); 596 thread_lock(td); 597 PROC_SUNLOCK(p); 598 599 /* Do the same timestamp bookkeeping that mi_switch() would do. */ 600 new_switchtime = cpu_ticks(); 601 runtime = new_switchtime - PCPU_GET(switchtime); 602 td->td_runtime += runtime; 603 td->td_incruntime += runtime; 604 PCPU_SET(switchtime, new_switchtime); 605 PCPU_SET(switchticks, ticks); 606 VM_CNT_INC(v_swtch); 607 608 /* Save our resource usage in our process. */ 609 td->td_ru.ru_nvcsw++; 610 ruxagg(p, td); 611 rucollect(&p->p_ru, &td->td_ru); 612 PROC_STATUNLOCK(p); 613 614 td->td_state = TDS_INACTIVE; 615 #ifdef WITNESS 616 witness_thread_exit(td); 617 #endif 618 CTR1(KTR_PROC, "thread_exit: cpu_throw() thread %p", td); 619 sched_throw(td); 620 panic("I'm a teapot!"); 621 /* NOTREACHED */ 622 } 623 624 /* 625 * Do any thread specific cleanups that may be needed in wait() 626 * called with Giant, proc and schedlock not held. 627 */ 628 void 629 thread_wait(struct proc *p) 630 { 631 struct thread *td; 632 633 mtx_assert(&Giant, MA_NOTOWNED); 634 KASSERT(p->p_numthreads == 1, ("multiple threads in thread_wait()")); 635 KASSERT(p->p_exitthreads == 0, ("p_exitthreads leaking")); 636 td = FIRST_THREAD_IN_PROC(p); 637 /* Lock the last thread so we spin until it exits cpu_throw(). */ 638 thread_lock(td); 639 thread_unlock(td); 640 lock_profile_thread_exit(td); 641 cpuset_rel(td->td_cpuset); 642 td->td_cpuset = NULL; 643 cpu_thread_clean(td); 644 thread_cow_free(td); 645 callout_drain(&td->td_slpcallout); 646 thread_reap(); /* check for zombie threads etc. */ 647 } 648 649 /* 650 * Link a thread to a process. 651 * set up anything that needs to be initialized for it to 652 * be used by the process. 653 */ 654 void 655 thread_link(struct thread *td, struct proc *p) 656 { 657 658 /* 659 * XXX This can't be enabled because it's called for proc0 before 660 * its lock has been created. 661 * PROC_LOCK_ASSERT(p, MA_OWNED); 662 */ 663 td->td_state = TDS_INACTIVE; 664 td->td_proc = p; 665 td->td_flags = TDF_INMEM; 666 667 LIST_INIT(&td->td_contested); 668 LIST_INIT(&td->td_lprof[0]); 669 LIST_INIT(&td->td_lprof[1]); 670 sigqueue_init(&td->td_sigqueue, p); 671 callout_init(&td->td_slpcallout, 1); 672 TAILQ_INSERT_TAIL(&p->p_threads, td, td_plist); 673 p->p_numthreads++; 674 } 675 676 /* 677 * Called from: 678 * thread_exit() 679 */ 680 void 681 thread_unlink(struct thread *td) 682 { 683 struct proc *p = td->td_proc; 684 685 PROC_LOCK_ASSERT(p, MA_OWNED); 686 TAILQ_REMOVE(&p->p_threads, td, td_plist); 687 p->p_numthreads--; 688 /* could clear a few other things here */ 689 /* Must NOT clear links to proc! */ 690 } 691 692 static int 693 calc_remaining(struct proc *p, int mode) 694 { 695 int remaining; 696 697 PROC_LOCK_ASSERT(p, MA_OWNED); 698 PROC_SLOCK_ASSERT(p, MA_OWNED); 699 if (mode == SINGLE_EXIT) 700 remaining = p->p_numthreads; 701 else if (mode == SINGLE_BOUNDARY) 702 remaining = p->p_numthreads - p->p_boundary_count; 703 else if (mode == SINGLE_NO_EXIT || mode == SINGLE_ALLPROC) 704 remaining = p->p_numthreads - p->p_suspcount; 705 else 706 panic("calc_remaining: wrong mode %d", mode); 707 return (remaining); 708 } 709 710 static int 711 remain_for_mode(int mode) 712 { 713 714 return (mode == SINGLE_ALLPROC ? 0 : 1); 715 } 716 717 static int 718 weed_inhib(int mode, struct thread *td2, struct proc *p) 719 { 720 int wakeup_swapper; 721 722 PROC_LOCK_ASSERT(p, MA_OWNED); 723 PROC_SLOCK_ASSERT(p, MA_OWNED); 724 THREAD_LOCK_ASSERT(td2, MA_OWNED); 725 726 wakeup_swapper = 0; 727 switch (mode) { 728 case SINGLE_EXIT: 729 if (TD_IS_SUSPENDED(td2)) 730 wakeup_swapper |= thread_unsuspend_one(td2, p, true); 731 if (TD_ON_SLEEPQ(td2) && (td2->td_flags & TDF_SINTR) != 0) 732 wakeup_swapper |= sleepq_abort(td2, EINTR); 733 break; 734 case SINGLE_BOUNDARY: 735 case SINGLE_NO_EXIT: 736 if (TD_IS_SUSPENDED(td2) && (td2->td_flags & TDF_BOUNDARY) == 0) 737 wakeup_swapper |= thread_unsuspend_one(td2, p, false); 738 if (TD_ON_SLEEPQ(td2) && (td2->td_flags & TDF_SINTR) != 0) 739 wakeup_swapper |= sleepq_abort(td2, ERESTART); 740 break; 741 case SINGLE_ALLPROC: 742 /* 743 * ALLPROC suspend tries to avoid spurious EINTR for 744 * threads sleeping interruptable, by suspending the 745 * thread directly, similarly to sig_suspend_threads(). 746 * Since such sleep is not performed at the user 747 * boundary, TDF_BOUNDARY flag is not set, and TDF_ALLPROCSUSP 748 * is used to avoid immediate un-suspend. 749 */ 750 if (TD_IS_SUSPENDED(td2) && (td2->td_flags & (TDF_BOUNDARY | 751 TDF_ALLPROCSUSP)) == 0) 752 wakeup_swapper |= thread_unsuspend_one(td2, p, false); 753 if (TD_ON_SLEEPQ(td2) && (td2->td_flags & TDF_SINTR) != 0) { 754 if ((td2->td_flags & TDF_SBDRY) == 0) { 755 thread_suspend_one(td2); 756 td2->td_flags |= TDF_ALLPROCSUSP; 757 } else { 758 wakeup_swapper |= sleepq_abort(td2, ERESTART); 759 } 760 } 761 break; 762 } 763 return (wakeup_swapper); 764 } 765 766 /* 767 * Enforce single-threading. 768 * 769 * Returns 1 if the caller must abort (another thread is waiting to 770 * exit the process or similar). Process is locked! 771 * Returns 0 when you are successfully the only thread running. 772 * A process has successfully single threaded in the suspend mode when 773 * There are no threads in user mode. Threads in the kernel must be 774 * allowed to continue until they get to the user boundary. They may even 775 * copy out their return values and data before suspending. They may however be 776 * accelerated in reaching the user boundary as we will wake up 777 * any sleeping threads that are interruptable. (PCATCH). 778 */ 779 int 780 thread_single(struct proc *p, int mode) 781 { 782 struct thread *td; 783 struct thread *td2; 784 int remaining, wakeup_swapper; 785 786 td = curthread; 787 KASSERT(mode == SINGLE_EXIT || mode == SINGLE_BOUNDARY || 788 mode == SINGLE_ALLPROC || mode == SINGLE_NO_EXIT, 789 ("invalid mode %d", mode)); 790 /* 791 * If allowing non-ALLPROC singlethreading for non-curproc 792 * callers, calc_remaining() and remain_for_mode() should be 793 * adjusted to also account for td->td_proc != p. For now 794 * this is not implemented because it is not used. 795 */ 796 KASSERT((mode == SINGLE_ALLPROC && td->td_proc != p) || 797 (mode != SINGLE_ALLPROC && td->td_proc == p), 798 ("mode %d proc %p curproc %p", mode, p, td->td_proc)); 799 mtx_assert(&Giant, MA_NOTOWNED); 800 PROC_LOCK_ASSERT(p, MA_OWNED); 801 802 if ((p->p_flag & P_HADTHREADS) == 0 && mode != SINGLE_ALLPROC) 803 return (0); 804 805 /* Is someone already single threading? */ 806 if (p->p_singlethread != NULL && p->p_singlethread != td) 807 return (1); 808 809 if (mode == SINGLE_EXIT) { 810 p->p_flag |= P_SINGLE_EXIT; 811 p->p_flag &= ~P_SINGLE_BOUNDARY; 812 } else { 813 p->p_flag &= ~P_SINGLE_EXIT; 814 if (mode == SINGLE_BOUNDARY) 815 p->p_flag |= P_SINGLE_BOUNDARY; 816 else 817 p->p_flag &= ~P_SINGLE_BOUNDARY; 818 } 819 if (mode == SINGLE_ALLPROC) 820 p->p_flag |= P_TOTAL_STOP; 821 p->p_flag |= P_STOPPED_SINGLE; 822 PROC_SLOCK(p); 823 p->p_singlethread = td; 824 remaining = calc_remaining(p, mode); 825 while (remaining != remain_for_mode(mode)) { 826 if (P_SHOULDSTOP(p) != P_STOPPED_SINGLE) 827 goto stopme; 828 wakeup_swapper = 0; 829 FOREACH_THREAD_IN_PROC(p, td2) { 830 if (td2 == td) 831 continue; 832 thread_lock(td2); 833 td2->td_flags |= TDF_ASTPENDING | TDF_NEEDSUSPCHK; 834 if (TD_IS_INHIBITED(td2)) { 835 wakeup_swapper |= weed_inhib(mode, td2, p); 836 #ifdef SMP 837 } else if (TD_IS_RUNNING(td2) && td != td2) { 838 forward_signal(td2); 839 #endif 840 } 841 thread_unlock(td2); 842 } 843 if (wakeup_swapper) 844 kick_proc0(); 845 remaining = calc_remaining(p, mode); 846 847 /* 848 * Maybe we suspended some threads.. was it enough? 849 */ 850 if (remaining == remain_for_mode(mode)) 851 break; 852 853 stopme: 854 /* 855 * Wake us up when everyone else has suspended. 856 * In the mean time we suspend as well. 857 */ 858 thread_suspend_switch(td, p); 859 remaining = calc_remaining(p, mode); 860 } 861 if (mode == SINGLE_EXIT) { 862 /* 863 * Convert the process to an unthreaded process. The 864 * SINGLE_EXIT is called by exit1() or execve(), in 865 * both cases other threads must be retired. 866 */ 867 KASSERT(p->p_numthreads == 1, ("Unthreading with >1 threads")); 868 p->p_singlethread = NULL; 869 p->p_flag &= ~(P_STOPPED_SINGLE | P_SINGLE_EXIT | P_HADTHREADS); 870 871 /* 872 * Wait for any remaining threads to exit cpu_throw(). 873 */ 874 while (p->p_exitthreads != 0) { 875 PROC_SUNLOCK(p); 876 PROC_UNLOCK(p); 877 sched_relinquish(td); 878 PROC_LOCK(p); 879 PROC_SLOCK(p); 880 } 881 } else if (mode == SINGLE_BOUNDARY) { 882 /* 883 * Wait until all suspended threads are removed from 884 * the processors. The thread_suspend_check() 885 * increments p_boundary_count while it is still 886 * running, which makes it possible for the execve() 887 * to destroy vmspace while our other threads are 888 * still using the address space. 889 * 890 * We lock the thread, which is only allowed to 891 * succeed after context switch code finished using 892 * the address space. 893 */ 894 FOREACH_THREAD_IN_PROC(p, td2) { 895 if (td2 == td) 896 continue; 897 thread_lock(td2); 898 KASSERT((td2->td_flags & TDF_BOUNDARY) != 0, 899 ("td %p not on boundary", td2)); 900 KASSERT(TD_IS_SUSPENDED(td2), 901 ("td %p is not suspended", td2)); 902 thread_unlock(td2); 903 } 904 } 905 PROC_SUNLOCK(p); 906 return (0); 907 } 908 909 bool 910 thread_suspend_check_needed(void) 911 { 912 struct proc *p; 913 struct thread *td; 914 915 td = curthread; 916 p = td->td_proc; 917 PROC_LOCK_ASSERT(p, MA_OWNED); 918 return (P_SHOULDSTOP(p) || ((p->p_flag & P_TRACED) != 0 && 919 (td->td_dbgflags & TDB_SUSPEND) != 0)); 920 } 921 922 /* 923 * Called in from locations that can safely check to see 924 * whether we have to suspend or at least throttle for a 925 * single-thread event (e.g. fork). 926 * 927 * Such locations include userret(). 928 * If the "return_instead" argument is non zero, the thread must be able to 929 * accept 0 (caller may continue), or 1 (caller must abort) as a result. 930 * 931 * The 'return_instead' argument tells the function if it may do a 932 * thread_exit() or suspend, or whether the caller must abort and back 933 * out instead. 934 * 935 * If the thread that set the single_threading request has set the 936 * P_SINGLE_EXIT bit in the process flags then this call will never return 937 * if 'return_instead' is false, but will exit. 938 * 939 * P_SINGLE_EXIT | return_instead == 0| return_instead != 0 940 *---------------+--------------------+--------------------- 941 * 0 | returns 0 | returns 0 or 1 942 * | when ST ends | immediately 943 *---------------+--------------------+--------------------- 944 * 1 | thread exits | returns 1 945 * | | immediately 946 * 0 = thread_exit() or suspension ok, 947 * other = return error instead of stopping the thread. 948 * 949 * While a full suspension is under effect, even a single threading 950 * thread would be suspended if it made this call (but it shouldn't). 951 * This call should only be made from places where 952 * thread_exit() would be safe as that may be the outcome unless 953 * return_instead is set. 954 */ 955 int 956 thread_suspend_check(int return_instead) 957 { 958 struct thread *td; 959 struct proc *p; 960 int wakeup_swapper; 961 962 td = curthread; 963 p = td->td_proc; 964 mtx_assert(&Giant, MA_NOTOWNED); 965 PROC_LOCK_ASSERT(p, MA_OWNED); 966 while (thread_suspend_check_needed()) { 967 if (P_SHOULDSTOP(p) == P_STOPPED_SINGLE) { 968 KASSERT(p->p_singlethread != NULL, 969 ("singlethread not set")); 970 /* 971 * The only suspension in action is a 972 * single-threading. Single threader need not stop. 973 * It is safe to access p->p_singlethread unlocked 974 * because it can only be set to our address by us. 975 */ 976 if (p->p_singlethread == td) 977 return (0); /* Exempt from stopping. */ 978 } 979 if ((p->p_flag & P_SINGLE_EXIT) && return_instead) 980 return (EINTR); 981 982 /* Should we goto user boundary if we didn't come from there? */ 983 if (P_SHOULDSTOP(p) == P_STOPPED_SINGLE && 984 (p->p_flag & P_SINGLE_BOUNDARY) && return_instead) 985 return (ERESTART); 986 987 /* 988 * Ignore suspend requests if they are deferred. 989 */ 990 if ((td->td_flags & TDF_SBDRY) != 0) { 991 KASSERT(return_instead, 992 ("TDF_SBDRY set for unsafe thread_suspend_check")); 993 KASSERT((td->td_flags & (TDF_SEINTR | TDF_SERESTART)) != 994 (TDF_SEINTR | TDF_SERESTART), 995 ("both TDF_SEINTR and TDF_SERESTART")); 996 return (TD_SBDRY_INTR(td) ? TD_SBDRY_ERRNO(td) : 0); 997 } 998 999 /* 1000 * If the process is waiting for us to exit, 1001 * this thread should just suicide. 1002 * Assumes that P_SINGLE_EXIT implies P_STOPPED_SINGLE. 1003 */ 1004 if ((p->p_flag & P_SINGLE_EXIT) && (p->p_singlethread != td)) { 1005 PROC_UNLOCK(p); 1006 1007 /* 1008 * Allow Linux emulation layer to do some work 1009 * before thread suicide. 1010 */ 1011 if (__predict_false(p->p_sysent->sv_thread_detach != NULL)) 1012 (p->p_sysent->sv_thread_detach)(td); 1013 umtx_thread_exit(td); 1014 kern_thr_exit(td); 1015 panic("stopped thread did not exit"); 1016 } 1017 1018 PROC_SLOCK(p); 1019 thread_stopped(p); 1020 if (P_SHOULDSTOP(p) == P_STOPPED_SINGLE) { 1021 if (p->p_numthreads == p->p_suspcount + 1) { 1022 thread_lock(p->p_singlethread); 1023 wakeup_swapper = thread_unsuspend_one( 1024 p->p_singlethread, p, false); 1025 thread_unlock(p->p_singlethread); 1026 if (wakeup_swapper) 1027 kick_proc0(); 1028 } 1029 } 1030 PROC_UNLOCK(p); 1031 thread_lock(td); 1032 /* 1033 * When a thread suspends, it just 1034 * gets taken off all queues. 1035 */ 1036 thread_suspend_one(td); 1037 if (return_instead == 0) { 1038 p->p_boundary_count++; 1039 td->td_flags |= TDF_BOUNDARY; 1040 } 1041 PROC_SUNLOCK(p); 1042 mi_switch(SW_INVOL | SWT_SUSPEND, NULL); 1043 thread_unlock(td); 1044 PROC_LOCK(p); 1045 } 1046 return (0); 1047 } 1048 1049 void 1050 thread_suspend_switch(struct thread *td, struct proc *p) 1051 { 1052 1053 KASSERT(!TD_IS_SUSPENDED(td), ("already suspended")); 1054 PROC_LOCK_ASSERT(p, MA_OWNED); 1055 PROC_SLOCK_ASSERT(p, MA_OWNED); 1056 /* 1057 * We implement thread_suspend_one in stages here to avoid 1058 * dropping the proc lock while the thread lock is owned. 1059 */ 1060 if (p == td->td_proc) { 1061 thread_stopped(p); 1062 p->p_suspcount++; 1063 } 1064 PROC_UNLOCK(p); 1065 thread_lock(td); 1066 td->td_flags &= ~TDF_NEEDSUSPCHK; 1067 TD_SET_SUSPENDED(td); 1068 sched_sleep(td, 0); 1069 PROC_SUNLOCK(p); 1070 DROP_GIANT(); 1071 mi_switch(SW_VOL | SWT_SUSPEND, NULL); 1072 thread_unlock(td); 1073 PICKUP_GIANT(); 1074 PROC_LOCK(p); 1075 PROC_SLOCK(p); 1076 } 1077 1078 void 1079 thread_suspend_one(struct thread *td) 1080 { 1081 struct proc *p; 1082 1083 p = td->td_proc; 1084 PROC_SLOCK_ASSERT(p, MA_OWNED); 1085 THREAD_LOCK_ASSERT(td, MA_OWNED); 1086 KASSERT(!TD_IS_SUSPENDED(td), ("already suspended")); 1087 p->p_suspcount++; 1088 td->td_flags &= ~TDF_NEEDSUSPCHK; 1089 TD_SET_SUSPENDED(td); 1090 sched_sleep(td, 0); 1091 } 1092 1093 static int 1094 thread_unsuspend_one(struct thread *td, struct proc *p, bool boundary) 1095 { 1096 1097 THREAD_LOCK_ASSERT(td, MA_OWNED); 1098 KASSERT(TD_IS_SUSPENDED(td), ("Thread not suspended")); 1099 TD_CLR_SUSPENDED(td); 1100 td->td_flags &= ~TDF_ALLPROCSUSP; 1101 if (td->td_proc == p) { 1102 PROC_SLOCK_ASSERT(p, MA_OWNED); 1103 p->p_suspcount--; 1104 if (boundary && (td->td_flags & TDF_BOUNDARY) != 0) { 1105 td->td_flags &= ~TDF_BOUNDARY; 1106 p->p_boundary_count--; 1107 } 1108 } 1109 return (setrunnable(td)); 1110 } 1111 1112 /* 1113 * Allow all threads blocked by single threading to continue running. 1114 */ 1115 void 1116 thread_unsuspend(struct proc *p) 1117 { 1118 struct thread *td; 1119 int wakeup_swapper; 1120 1121 PROC_LOCK_ASSERT(p, MA_OWNED); 1122 PROC_SLOCK_ASSERT(p, MA_OWNED); 1123 wakeup_swapper = 0; 1124 if (!P_SHOULDSTOP(p)) { 1125 FOREACH_THREAD_IN_PROC(p, td) { 1126 thread_lock(td); 1127 if (TD_IS_SUSPENDED(td)) { 1128 wakeup_swapper |= thread_unsuspend_one(td, p, 1129 true); 1130 } 1131 thread_unlock(td); 1132 } 1133 } else if (P_SHOULDSTOP(p) == P_STOPPED_SINGLE && 1134 p->p_numthreads == p->p_suspcount) { 1135 /* 1136 * Stopping everything also did the job for the single 1137 * threading request. Now we've downgraded to single-threaded, 1138 * let it continue. 1139 */ 1140 if (p->p_singlethread->td_proc == p) { 1141 thread_lock(p->p_singlethread); 1142 wakeup_swapper = thread_unsuspend_one( 1143 p->p_singlethread, p, false); 1144 thread_unlock(p->p_singlethread); 1145 } 1146 } 1147 if (wakeup_swapper) 1148 kick_proc0(); 1149 } 1150 1151 /* 1152 * End the single threading mode.. 1153 */ 1154 void 1155 thread_single_end(struct proc *p, int mode) 1156 { 1157 struct thread *td; 1158 int wakeup_swapper; 1159 1160 KASSERT(mode == SINGLE_EXIT || mode == SINGLE_BOUNDARY || 1161 mode == SINGLE_ALLPROC || mode == SINGLE_NO_EXIT, 1162 ("invalid mode %d", mode)); 1163 PROC_LOCK_ASSERT(p, MA_OWNED); 1164 KASSERT((mode == SINGLE_ALLPROC && (p->p_flag & P_TOTAL_STOP) != 0) || 1165 (mode != SINGLE_ALLPROC && (p->p_flag & P_TOTAL_STOP) == 0), 1166 ("mode %d does not match P_TOTAL_STOP", mode)); 1167 KASSERT(mode == SINGLE_ALLPROC || p->p_singlethread == curthread, 1168 ("thread_single_end from other thread %p %p", 1169 curthread, p->p_singlethread)); 1170 KASSERT(mode != SINGLE_BOUNDARY || 1171 (p->p_flag & P_SINGLE_BOUNDARY) != 0, 1172 ("mis-matched SINGLE_BOUNDARY flags %x", p->p_flag)); 1173 p->p_flag &= ~(P_STOPPED_SINGLE | P_SINGLE_EXIT | P_SINGLE_BOUNDARY | 1174 P_TOTAL_STOP); 1175 PROC_SLOCK(p); 1176 p->p_singlethread = NULL; 1177 wakeup_swapper = 0; 1178 /* 1179 * If there are other threads they may now run, 1180 * unless of course there is a blanket 'stop order' 1181 * on the process. The single threader must be allowed 1182 * to continue however as this is a bad place to stop. 1183 */ 1184 if (p->p_numthreads != remain_for_mode(mode) && !P_SHOULDSTOP(p)) { 1185 FOREACH_THREAD_IN_PROC(p, td) { 1186 thread_lock(td); 1187 if (TD_IS_SUSPENDED(td)) { 1188 wakeup_swapper |= thread_unsuspend_one(td, p, 1189 mode == SINGLE_BOUNDARY); 1190 } 1191 thread_unlock(td); 1192 } 1193 } 1194 KASSERT(mode != SINGLE_BOUNDARY || p->p_boundary_count == 0, 1195 ("inconsistent boundary count %d", p->p_boundary_count)); 1196 PROC_SUNLOCK(p); 1197 if (wakeup_swapper) 1198 kick_proc0(); 1199 } 1200 1201 struct thread * 1202 thread_find(struct proc *p, lwpid_t tid) 1203 { 1204 struct thread *td; 1205 1206 PROC_LOCK_ASSERT(p, MA_OWNED); 1207 FOREACH_THREAD_IN_PROC(p, td) { 1208 if (td->td_tid == tid) 1209 break; 1210 } 1211 return (td); 1212 } 1213 1214 /* Locate a thread by number; return with proc lock held. */ 1215 struct thread * 1216 tdfind(lwpid_t tid, pid_t pid) 1217 { 1218 #define RUN_THRESH 16 1219 struct thread *td; 1220 int run = 0; 1221 1222 rw_rlock(&tidhash_lock); 1223 LIST_FOREACH(td, TIDHASH(tid), td_hash) { 1224 if (td->td_tid == tid) { 1225 if (pid != -1 && td->td_proc->p_pid != pid) { 1226 td = NULL; 1227 break; 1228 } 1229 PROC_LOCK(td->td_proc); 1230 if (td->td_proc->p_state == PRS_NEW) { 1231 PROC_UNLOCK(td->td_proc); 1232 td = NULL; 1233 break; 1234 } 1235 if (run > RUN_THRESH) { 1236 if (rw_try_upgrade(&tidhash_lock)) { 1237 LIST_REMOVE(td, td_hash); 1238 LIST_INSERT_HEAD(TIDHASH(td->td_tid), 1239 td, td_hash); 1240 rw_wunlock(&tidhash_lock); 1241 return (td); 1242 } 1243 } 1244 break; 1245 } 1246 run++; 1247 } 1248 rw_runlock(&tidhash_lock); 1249 return (td); 1250 } 1251 1252 void 1253 tidhash_add(struct thread *td) 1254 { 1255 rw_wlock(&tidhash_lock); 1256 LIST_INSERT_HEAD(TIDHASH(td->td_tid), td, td_hash); 1257 rw_wunlock(&tidhash_lock); 1258 } 1259 1260 void 1261 tidhash_remove(struct thread *td) 1262 { 1263 rw_wlock(&tidhash_lock); 1264 LIST_REMOVE(td, td_hash); 1265 rw_wunlock(&tidhash_lock); 1266 } 1267