1 /*- 2 * SPDX-License-Identifier: BSD-2-Clause 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/msan.h> 42 #include <sys/mutex.h> 43 #include <sys/proc.h> 44 #include <sys/bitstring.h> 45 #include <sys/epoch.h> 46 #include <sys/rangelock.h> 47 #include <sys/resourcevar.h> 48 #include <sys/sdt.h> 49 #include <sys/smp.h> 50 #include <sys/sched.h> 51 #include <sys/sleepqueue.h> 52 #include <sys/selinfo.h> 53 #include <sys/syscallsubr.h> 54 #include <sys/dtrace_bsd.h> 55 #include <sys/sysent.h> 56 #include <sys/turnstile.h> 57 #include <sys/taskqueue.h> 58 #include <sys/ktr.h> 59 #include <sys/rwlock.h> 60 #include <sys/umtxvar.h> 61 #include <sys/vmmeter.h> 62 #include <sys/cpuset.h> 63 #ifdef HWPMC_HOOKS 64 #include <sys/pmckern.h> 65 #endif 66 #include <sys/priv.h> 67 68 #include <security/audit/audit.h> 69 70 #include <vm/pmap.h> 71 #include <vm/vm.h> 72 #include <vm/vm_extern.h> 73 #include <vm/uma.h> 74 #include <vm/vm_phys.h> 75 #include <sys/eventhandler.h> 76 77 /* 78 * Asserts below verify the stability of struct thread and struct proc 79 * layout, as exposed by KBI to modules. On head, the KBI is allowed 80 * to drift, change to the structures must be accompanied by the 81 * assert update. 82 * 83 * On the stable branches after KBI freeze, conditions must not be 84 * violated. Typically new fields are moved to the end of the 85 * structures. 86 */ 87 #ifdef __amd64__ 88 _Static_assert(offsetof(struct thread, td_flags) == 0x108, 89 "struct thread KBI td_flags"); 90 _Static_assert(offsetof(struct thread, td_pflags) == 0x114, 91 "struct thread KBI td_pflags"); 92 _Static_assert(offsetof(struct thread, td_frame) == 0x4b8, 93 "struct thread KBI td_frame"); 94 _Static_assert(offsetof(struct thread, td_emuldata) == 0x6c0, 95 "struct thread KBI td_emuldata"); 96 _Static_assert(offsetof(struct proc, p_flag) == 0xb8, 97 "struct proc KBI p_flag"); 98 _Static_assert(offsetof(struct proc, p_pid) == 0xc4, 99 "struct proc KBI p_pid"); 100 _Static_assert(offsetof(struct proc, p_filemon) == 0x3c8, 101 "struct proc KBI p_filemon"); 102 _Static_assert(offsetof(struct proc, p_comm) == 0x3e0, 103 "struct proc KBI p_comm"); 104 _Static_assert(offsetof(struct proc, p_emuldata) == 0x4d0, 105 "struct proc KBI p_emuldata"); 106 #endif 107 #ifdef __i386__ 108 _Static_assert(offsetof(struct thread, td_flags) == 0x9c, 109 "struct thread KBI td_flags"); 110 _Static_assert(offsetof(struct thread, td_pflags) == 0xa8, 111 "struct thread KBI td_pflags"); 112 _Static_assert(offsetof(struct thread, td_frame) == 0x314, 113 "struct thread KBI td_frame"); 114 _Static_assert(offsetof(struct thread, td_emuldata) == 0x358, 115 "struct thread KBI td_emuldata"); 116 _Static_assert(offsetof(struct proc, p_flag) == 0x6c, 117 "struct proc KBI p_flag"); 118 _Static_assert(offsetof(struct proc, p_pid) == 0x78, 119 "struct proc KBI p_pid"); 120 _Static_assert(offsetof(struct proc, p_filemon) == 0x270, 121 "struct proc KBI p_filemon"); 122 _Static_assert(offsetof(struct proc, p_comm) == 0x284, 123 "struct proc KBI p_comm"); 124 _Static_assert(offsetof(struct proc, p_emuldata) == 0x318, 125 "struct proc KBI p_emuldata"); 126 #endif 127 128 SDT_PROVIDER_DECLARE(proc); 129 SDT_PROBE_DEFINE(proc, , , lwp__exit); 130 131 /* 132 * thread related storage. 133 */ 134 static uma_zone_t thread_zone; 135 136 struct thread_domain_data { 137 struct thread *tdd_zombies; 138 int tdd_reapticks; 139 } __aligned(CACHE_LINE_SIZE); 140 141 static struct thread_domain_data thread_domain_data[MAXMEMDOM]; 142 143 static struct task thread_reap_task; 144 static struct callout thread_reap_callout; 145 146 static void thread_zombie(struct thread *); 147 static void thread_reap(void); 148 static void thread_reap_all(void); 149 static void thread_reap_task_cb(void *, int); 150 static void thread_reap_callout_cb(void *); 151 static int thread_unsuspend_one(struct thread *td, struct proc *p, 152 bool boundary); 153 static void thread_free_batched(struct thread *td); 154 155 static __exclusive_cache_line struct mtx tid_lock; 156 static bitstr_t *tid_bitmap; 157 158 static MALLOC_DEFINE(M_TIDHASH, "tidhash", "thread hash"); 159 160 static int maxthread; 161 SYSCTL_INT(_kern, OID_AUTO, maxthread, CTLFLAG_RDTUN, 162 &maxthread, 0, "Maximum number of threads"); 163 164 static __exclusive_cache_line int nthreads; 165 166 static LIST_HEAD(tidhashhead, thread) *tidhashtbl; 167 static u_long tidhash; 168 static u_long tidhashlock; 169 static struct rwlock *tidhashtbl_lock; 170 #define TIDHASH(tid) (&tidhashtbl[(tid) & tidhash]) 171 #define TIDHASHLOCK(tid) (&tidhashtbl_lock[(tid) & tidhashlock]) 172 173 EVENTHANDLER_LIST_DEFINE(thread_ctor); 174 EVENTHANDLER_LIST_DEFINE(thread_dtor); 175 EVENTHANDLER_LIST_DEFINE(thread_init); 176 EVENTHANDLER_LIST_DEFINE(thread_fini); 177 178 static bool 179 thread_count_inc_try(void) 180 { 181 int nthreads_new; 182 183 nthreads_new = atomic_fetchadd_int(&nthreads, 1) + 1; 184 if (nthreads_new >= maxthread - 100) { 185 if (priv_check_cred(curthread->td_ucred, PRIV_MAXPROC) != 0 || 186 nthreads_new >= maxthread) { 187 atomic_subtract_int(&nthreads, 1); 188 return (false); 189 } 190 } 191 return (true); 192 } 193 194 static bool 195 thread_count_inc(void) 196 { 197 static struct timeval lastfail; 198 static int curfail; 199 200 thread_reap(); 201 if (thread_count_inc_try()) { 202 return (true); 203 } 204 205 thread_reap_all(); 206 if (thread_count_inc_try()) { 207 return (true); 208 } 209 210 if (ppsratecheck(&lastfail, &curfail, 1)) { 211 printf("maxthread limit exceeded by uid %u " 212 "(pid %d); consider increasing kern.maxthread\n", 213 curthread->td_ucred->cr_ruid, curproc->p_pid); 214 } 215 return (false); 216 } 217 218 static void 219 thread_count_sub(int n) 220 { 221 222 atomic_subtract_int(&nthreads, n); 223 } 224 225 static void 226 thread_count_dec(void) 227 { 228 229 thread_count_sub(1); 230 } 231 232 static lwpid_t 233 tid_alloc(void) 234 { 235 static lwpid_t trytid; 236 lwpid_t tid; 237 238 mtx_lock(&tid_lock); 239 /* 240 * It is an invariant that the bitmap is big enough to hold maxthread 241 * IDs. If we got to this point there has to be at least one free. 242 */ 243 if (trytid >= maxthread) 244 trytid = 0; 245 bit_ffc_at(tid_bitmap, trytid, maxthread, &tid); 246 if (tid == -1) { 247 KASSERT(trytid != 0, ("unexpectedly ran out of IDs")); 248 trytid = 0; 249 bit_ffc_at(tid_bitmap, trytid, maxthread, &tid); 250 KASSERT(tid != -1, ("unexpectedly ran out of IDs")); 251 } 252 bit_set(tid_bitmap, tid); 253 trytid = tid + 1; 254 mtx_unlock(&tid_lock); 255 return (tid + NO_PID); 256 } 257 258 static void 259 tid_free_locked(lwpid_t rtid) 260 { 261 lwpid_t tid; 262 263 mtx_assert(&tid_lock, MA_OWNED); 264 KASSERT(rtid >= NO_PID, 265 ("%s: invalid tid %d\n", __func__, rtid)); 266 tid = rtid - NO_PID; 267 KASSERT(bit_test(tid_bitmap, tid) != 0, 268 ("thread ID %d not allocated\n", rtid)); 269 bit_clear(tid_bitmap, tid); 270 } 271 272 static void 273 tid_free(lwpid_t rtid) 274 { 275 276 mtx_lock(&tid_lock); 277 tid_free_locked(rtid); 278 mtx_unlock(&tid_lock); 279 } 280 281 static void 282 tid_free_batch(lwpid_t *batch, int n) 283 { 284 int i; 285 286 mtx_lock(&tid_lock); 287 for (i = 0; i < n; i++) { 288 tid_free_locked(batch[i]); 289 } 290 mtx_unlock(&tid_lock); 291 } 292 293 /* 294 * Batching for thread reapping. 295 */ 296 struct tidbatch { 297 lwpid_t tab[16]; 298 int n; 299 }; 300 301 static void 302 tidbatch_prep(struct tidbatch *tb) 303 { 304 305 tb->n = 0; 306 } 307 308 static void 309 tidbatch_add(struct tidbatch *tb, struct thread *td) 310 { 311 312 KASSERT(tb->n < nitems(tb->tab), 313 ("%s: count too high %d", __func__, tb->n)); 314 tb->tab[tb->n] = td->td_tid; 315 tb->n++; 316 } 317 318 static void 319 tidbatch_process(struct tidbatch *tb) 320 { 321 322 KASSERT(tb->n <= nitems(tb->tab), 323 ("%s: count too high %d", __func__, tb->n)); 324 if (tb->n == nitems(tb->tab)) { 325 tid_free_batch(tb->tab, tb->n); 326 tb->n = 0; 327 } 328 } 329 330 static void 331 tidbatch_final(struct tidbatch *tb) 332 { 333 334 KASSERT(tb->n <= nitems(tb->tab), 335 ("%s: count too high %d", __func__, tb->n)); 336 if (tb->n != 0) { 337 tid_free_batch(tb->tab, tb->n); 338 } 339 } 340 341 /* 342 * Batching thread count free, for consistency 343 */ 344 struct tdcountbatch { 345 int n; 346 }; 347 348 static void 349 tdcountbatch_prep(struct tdcountbatch *tb) 350 { 351 352 tb->n = 0; 353 } 354 355 static void 356 tdcountbatch_add(struct tdcountbatch *tb, struct thread *td __unused) 357 { 358 359 tb->n++; 360 } 361 362 static void 363 tdcountbatch_process(struct tdcountbatch *tb) 364 { 365 366 if (tb->n == 32) { 367 thread_count_sub(tb->n); 368 tb->n = 0; 369 } 370 } 371 372 static void 373 tdcountbatch_final(struct tdcountbatch *tb) 374 { 375 376 if (tb->n != 0) { 377 thread_count_sub(tb->n); 378 } 379 } 380 381 /* 382 * Prepare a thread for use. 383 */ 384 static int 385 thread_ctor(void *mem, int size, void *arg, int flags) 386 { 387 struct thread *td; 388 389 td = (struct thread *)mem; 390 TD_SET_STATE(td, TDS_INACTIVE); 391 td->td_lastcpu = td->td_oncpu = NOCPU; 392 393 /* 394 * Note that td_critnest begins life as 1 because the thread is not 395 * running and is thereby implicitly waiting to be on the receiving 396 * end of a context switch. 397 */ 398 td->td_critnest = 1; 399 td->td_lend_user_pri = PRI_MAX; 400 #ifdef AUDIT 401 audit_thread_alloc(td); 402 #endif 403 #ifdef KDTRACE_HOOKS 404 kdtrace_thread_ctor(td); 405 #endif 406 umtx_thread_alloc(td); 407 MPASS(td->td_sel == NULL); 408 return (0); 409 } 410 411 /* 412 * Reclaim a thread after use. 413 */ 414 static void 415 thread_dtor(void *mem, int size, void *arg) 416 { 417 struct thread *td; 418 419 td = (struct thread *)mem; 420 421 #ifdef INVARIANTS 422 /* Verify that this thread is in a safe state to free. */ 423 switch (TD_GET_STATE(td)) { 424 case TDS_INHIBITED: 425 case TDS_RUNNING: 426 case TDS_CAN_RUN: 427 case TDS_RUNQ: 428 /* 429 * We must never unlink a thread that is in one of 430 * these states, because it is currently active. 431 */ 432 panic("bad state for thread unlinking"); 433 /* NOTREACHED */ 434 case TDS_INACTIVE: 435 break; 436 default: 437 panic("bad thread state"); 438 /* NOTREACHED */ 439 } 440 #endif 441 #ifdef AUDIT 442 audit_thread_free(td); 443 #endif 444 #ifdef KDTRACE_HOOKS 445 kdtrace_thread_dtor(td); 446 #endif 447 /* Free all OSD associated to this thread. */ 448 osd_thread_exit(td); 449 ast_kclear(td); 450 seltdfini(td); 451 } 452 453 /* 454 * Initialize type-stable parts of a thread (when newly created). 455 */ 456 static int 457 thread_init(void *mem, int size, int flags) 458 { 459 struct thread *td; 460 461 td = (struct thread *)mem; 462 463 td->td_allocdomain = vm_phys_domain(vtophys(td)); 464 td->td_sleepqueue = sleepq_alloc(); 465 td->td_turnstile = turnstile_alloc(); 466 td->td_rlqe = NULL; 467 EVENTHANDLER_DIRECT_INVOKE(thread_init, td); 468 umtx_thread_init(td); 469 td->td_kstack = 0; 470 td->td_sel = NULL; 471 return (0); 472 } 473 474 /* 475 * Tear down type-stable parts of a thread (just before being discarded). 476 */ 477 static void 478 thread_fini(void *mem, int size) 479 { 480 struct thread *td; 481 482 td = (struct thread *)mem; 483 EVENTHANDLER_DIRECT_INVOKE(thread_fini, td); 484 rlqentry_free(td->td_rlqe); 485 turnstile_free(td->td_turnstile); 486 sleepq_free(td->td_sleepqueue); 487 umtx_thread_fini(td); 488 MPASS(td->td_sel == NULL); 489 } 490 491 /* 492 * For a newly created process, 493 * link up all the structures and its initial threads etc. 494 * called from: 495 * {arch}/{arch}/machdep.c {arch}_init(), init386() etc. 496 * proc_dtor() (should go away) 497 * proc_init() 498 */ 499 void 500 proc_linkup0(struct proc *p, struct thread *td) 501 { 502 TAILQ_INIT(&p->p_threads); /* all threads in proc */ 503 proc_linkup(p, td); 504 } 505 506 void 507 proc_linkup(struct proc *p, struct thread *td) 508 { 509 510 sigqueue_init(&p->p_sigqueue, p); 511 p->p_ksi = ksiginfo_alloc(M_WAITOK); 512 if (p->p_ksi != NULL) { 513 /* XXX p_ksi may be null if ksiginfo zone is not ready */ 514 p->p_ksi->ksi_flags = KSI_EXT | KSI_INS; 515 } 516 LIST_INIT(&p->p_mqnotifier); 517 p->p_numthreads = 0; 518 thread_link(td, p); 519 } 520 521 static void 522 ast_suspend(struct thread *td, int tda __unused) 523 { 524 struct proc *p; 525 526 p = td->td_proc; 527 /* 528 * We need to check to see if we have to exit or wait due to a 529 * single threading requirement or some other STOP condition. 530 */ 531 PROC_LOCK(p); 532 thread_suspend_check(0); 533 PROC_UNLOCK(p); 534 } 535 536 extern int max_threads_per_proc; 537 538 /* 539 * Initialize global thread allocation resources. 540 */ 541 void 542 threadinit(void) 543 { 544 u_long i; 545 lwpid_t tid0; 546 547 /* 548 * Place an upper limit on threads which can be allocated. 549 * 550 * Note that other factors may make the de facto limit much lower. 551 * 552 * Platform limits are somewhat arbitrary but deemed "more than good 553 * enough" for the foreseable future. 554 */ 555 if (maxthread == 0) { 556 #ifdef _LP64 557 maxthread = MIN(maxproc * max_threads_per_proc, 1000000); 558 #else 559 maxthread = MIN(maxproc * max_threads_per_proc, 100000); 560 #endif 561 } 562 563 mtx_init(&tid_lock, "TID lock", NULL, MTX_DEF); 564 tid_bitmap = bit_alloc(maxthread, M_TIDHASH, M_WAITOK); 565 /* 566 * Handle thread0. 567 */ 568 thread_count_inc(); 569 tid0 = tid_alloc(); 570 if (tid0 != THREAD0_TID) 571 panic("tid0 %d != %d\n", tid0, THREAD0_TID); 572 573 thread_zone = uma_zcreate("THREAD", sched_sizeof_thread(), 574 thread_ctor, thread_dtor, thread_init, thread_fini, 575 32 - 1, UMA_ZONE_NOFREE); 576 tidhashtbl = hashinit(maxproc / 2, M_TIDHASH, &tidhash); 577 tidhashlock = (tidhash + 1) / 64; 578 if (tidhashlock > 0) 579 tidhashlock--; 580 tidhashtbl_lock = malloc(sizeof(*tidhashtbl_lock) * (tidhashlock + 1), 581 M_TIDHASH, M_WAITOK | M_ZERO); 582 for (i = 0; i < tidhashlock + 1; i++) 583 rw_init(&tidhashtbl_lock[i], "tidhash"); 584 585 TASK_INIT(&thread_reap_task, 0, thread_reap_task_cb, NULL); 586 callout_init(&thread_reap_callout, 1); 587 callout_reset(&thread_reap_callout, 5 * hz, 588 thread_reap_callout_cb, NULL); 589 ast_register(TDA_SUSPEND, ASTR_ASTF_REQUIRED, 0, ast_suspend); 590 } 591 592 /* 593 * Place an unused thread on the zombie list. 594 */ 595 void 596 thread_zombie(struct thread *td) 597 { 598 struct thread_domain_data *tdd; 599 struct thread *ztd; 600 601 tdd = &thread_domain_data[td->td_allocdomain]; 602 ztd = atomic_load_ptr(&tdd->tdd_zombies); 603 for (;;) { 604 td->td_zombie = ztd; 605 if (atomic_fcmpset_rel_ptr((uintptr_t *)&tdd->tdd_zombies, 606 (uintptr_t *)&ztd, (uintptr_t)td)) 607 break; 608 continue; 609 } 610 } 611 612 /* 613 * Release a thread that has exited after cpu_throw(). 614 */ 615 void 616 thread_stash(struct thread *td) 617 { 618 atomic_subtract_rel_int(&td->td_proc->p_exitthreads, 1); 619 thread_zombie(td); 620 } 621 622 /* 623 * Reap zombies from passed domain. 624 */ 625 static void 626 thread_reap_domain(struct thread_domain_data *tdd) 627 { 628 struct thread *itd, *ntd; 629 struct tidbatch tidbatch; 630 struct credbatch credbatch; 631 struct limbatch limbatch; 632 struct tdcountbatch tdcountbatch; 633 634 /* 635 * Reading upfront is pessimal if followed by concurrent atomic_swap, 636 * but most of the time the list is empty. 637 */ 638 if (tdd->tdd_zombies == NULL) 639 return; 640 641 itd = (struct thread *)atomic_swap_ptr((uintptr_t *)&tdd->tdd_zombies, 642 (uintptr_t)NULL); 643 if (itd == NULL) 644 return; 645 646 /* 647 * Multiple CPUs can get here, the race is fine as ticks is only 648 * advisory. 649 */ 650 tdd->tdd_reapticks = ticks; 651 652 tidbatch_prep(&tidbatch); 653 credbatch_prep(&credbatch); 654 limbatch_prep(&limbatch); 655 tdcountbatch_prep(&tdcountbatch); 656 657 while (itd != NULL) { 658 ntd = itd->td_zombie; 659 EVENTHANDLER_DIRECT_INVOKE(thread_dtor, itd); 660 661 tidbatch_add(&tidbatch, itd); 662 credbatch_add(&credbatch, itd); 663 limbatch_add(&limbatch, itd); 664 tdcountbatch_add(&tdcountbatch, itd); 665 666 thread_free_batched(itd); 667 668 tidbatch_process(&tidbatch); 669 credbatch_process(&credbatch); 670 limbatch_process(&limbatch); 671 tdcountbatch_process(&tdcountbatch); 672 673 itd = ntd; 674 } 675 676 tidbatch_final(&tidbatch); 677 credbatch_final(&credbatch); 678 limbatch_final(&limbatch); 679 tdcountbatch_final(&tdcountbatch); 680 } 681 682 /* 683 * Reap zombies from all domains. 684 */ 685 static void 686 thread_reap_all(void) 687 { 688 struct thread_domain_data *tdd; 689 int i, domain; 690 691 domain = PCPU_GET(domain); 692 for (i = 0; i < vm_ndomains; i++) { 693 tdd = &thread_domain_data[(i + domain) % vm_ndomains]; 694 thread_reap_domain(tdd); 695 } 696 } 697 698 /* 699 * Reap zombies from local domain. 700 */ 701 static void 702 thread_reap(void) 703 { 704 struct thread_domain_data *tdd; 705 int domain; 706 707 domain = PCPU_GET(domain); 708 tdd = &thread_domain_data[domain]; 709 710 thread_reap_domain(tdd); 711 } 712 713 static void 714 thread_reap_task_cb(void *arg __unused, int pending __unused) 715 { 716 717 thread_reap_all(); 718 } 719 720 static void 721 thread_reap_callout_cb(void *arg __unused) 722 { 723 struct thread_domain_data *tdd; 724 int i, cticks, lticks; 725 bool wantreap; 726 727 wantreap = false; 728 cticks = atomic_load_int(&ticks); 729 for (i = 0; i < vm_ndomains; i++) { 730 tdd = &thread_domain_data[i]; 731 lticks = tdd->tdd_reapticks; 732 if (tdd->tdd_zombies != NULL && 733 (u_int)(cticks - lticks) > 5 * hz) { 734 wantreap = true; 735 break; 736 } 737 } 738 739 if (wantreap) 740 taskqueue_enqueue(taskqueue_thread, &thread_reap_task); 741 callout_reset(&thread_reap_callout, 5 * hz, 742 thread_reap_callout_cb, NULL); 743 } 744 745 /* 746 * Calling this function guarantees that any thread that exited before 747 * the call is reaped when the function returns. By 'exited' we mean 748 * a thread removed from the process linkage with thread_unlink(). 749 * Practically this means that caller must lock/unlock corresponding 750 * process lock before the call, to synchronize with thread_exit(). 751 */ 752 void 753 thread_reap_barrier(void) 754 { 755 struct task *t; 756 757 /* 758 * First do context switches to each CPU to ensure that all 759 * PCPU pc_deadthreads are moved to zombie list. 760 */ 761 quiesce_all_cpus("", PDROP); 762 763 /* 764 * Second, fire the task in the same thread as normal 765 * thread_reap() is done, to serialize reaping. 766 */ 767 t = malloc(sizeof(*t), M_TEMP, M_WAITOK); 768 TASK_INIT(t, 0, thread_reap_task_cb, t); 769 taskqueue_enqueue(taskqueue_thread, t); 770 taskqueue_drain(taskqueue_thread, t); 771 free(t, M_TEMP); 772 } 773 774 /* 775 * Allocate a thread. 776 */ 777 struct thread * 778 thread_alloc(int pages) 779 { 780 struct thread *td; 781 lwpid_t tid; 782 783 if (!thread_count_inc()) { 784 return (NULL); 785 } 786 787 tid = tid_alloc(); 788 td = uma_zalloc(thread_zone, M_WAITOK); 789 KASSERT(td->td_kstack == 0, ("thread_alloc got thread with kstack")); 790 if (!vm_thread_new(td, pages)) { 791 uma_zfree(thread_zone, td); 792 tid_free(tid); 793 thread_count_dec(); 794 return (NULL); 795 } 796 td->td_tid = tid; 797 bzero(&td->td_sa.args, sizeof(td->td_sa.args)); 798 kmsan_thread_alloc(td); 799 cpu_thread_alloc(td); 800 EVENTHANDLER_DIRECT_INVOKE(thread_ctor, td); 801 return (td); 802 } 803 804 int 805 thread_alloc_stack(struct thread *td, int pages) 806 { 807 808 KASSERT(td->td_kstack == 0, 809 ("thread_alloc_stack called on a thread with kstack")); 810 if (!vm_thread_new(td, pages)) 811 return (0); 812 cpu_thread_alloc(td); 813 return (1); 814 } 815 816 /* 817 * Deallocate a thread. 818 */ 819 static void 820 thread_free_batched(struct thread *td) 821 { 822 823 lock_profile_thread_exit(td); 824 if (td->td_cpuset) 825 cpuset_rel(td->td_cpuset); 826 td->td_cpuset = NULL; 827 cpu_thread_free(td); 828 if (td->td_kstack != 0) 829 vm_thread_dispose(td); 830 callout_drain(&td->td_slpcallout); 831 /* 832 * Freeing handled by the caller. 833 */ 834 td->td_tid = -1; 835 kmsan_thread_free(td); 836 uma_zfree(thread_zone, td); 837 } 838 839 void 840 thread_free(struct thread *td) 841 { 842 lwpid_t tid; 843 844 EVENTHANDLER_DIRECT_INVOKE(thread_dtor, td); 845 tid = td->td_tid; 846 thread_free_batched(td); 847 tid_free(tid); 848 thread_count_dec(); 849 } 850 851 void 852 thread_cow_get_proc(struct thread *newtd, struct proc *p) 853 { 854 855 PROC_LOCK_ASSERT(p, MA_OWNED); 856 newtd->td_realucred = crcowget(p->p_ucred); 857 newtd->td_ucred = newtd->td_realucred; 858 newtd->td_limit = lim_hold(p->p_limit); 859 newtd->td_cowgen = p->p_cowgen; 860 } 861 862 void 863 thread_cow_get(struct thread *newtd, struct thread *td) 864 { 865 866 MPASS(td->td_realucred == td->td_ucred); 867 newtd->td_realucred = crcowget(td->td_realucred); 868 newtd->td_ucred = newtd->td_realucred; 869 newtd->td_limit = lim_hold(td->td_limit); 870 newtd->td_cowgen = td->td_cowgen; 871 } 872 873 void 874 thread_cow_free(struct thread *td) 875 { 876 877 if (td->td_realucred != NULL) 878 crcowfree(td); 879 if (td->td_limit != NULL) 880 lim_free(td->td_limit); 881 } 882 883 void 884 thread_cow_update(struct thread *td) 885 { 886 struct proc *p; 887 struct ucred *oldcred; 888 struct plimit *oldlimit; 889 890 p = td->td_proc; 891 PROC_LOCK(p); 892 oldcred = crcowsync(); 893 oldlimit = lim_cowsync(); 894 td->td_cowgen = p->p_cowgen; 895 PROC_UNLOCK(p); 896 if (oldcred != NULL) 897 crfree(oldcred); 898 if (oldlimit != NULL) 899 lim_free(oldlimit); 900 } 901 902 void 903 thread_cow_synced(struct thread *td) 904 { 905 struct proc *p; 906 907 p = td->td_proc; 908 PROC_LOCK_ASSERT(p, MA_OWNED); 909 MPASS(td->td_cowgen != p->p_cowgen); 910 MPASS(td->td_ucred == p->p_ucred); 911 MPASS(td->td_limit == p->p_limit); 912 td->td_cowgen = p->p_cowgen; 913 } 914 915 /* 916 * Discard the current thread and exit from its context. 917 * Always called with scheduler locked. 918 * 919 * Because we can't free a thread while we're operating under its context, 920 * push the current thread into our CPU's deadthread holder. This means 921 * we needn't worry about someone else grabbing our context before we 922 * do a cpu_throw(). 923 */ 924 void 925 thread_exit(void) 926 { 927 uint64_t runtime, new_switchtime; 928 struct thread *td; 929 struct thread *td2; 930 struct proc *p; 931 int wakeup_swapper; 932 933 td = curthread; 934 p = td->td_proc; 935 936 PROC_SLOCK_ASSERT(p, MA_OWNED); 937 mtx_assert(&Giant, MA_NOTOWNED); 938 939 PROC_LOCK_ASSERT(p, MA_OWNED); 940 KASSERT(p != NULL, ("thread exiting without a process")); 941 CTR3(KTR_PROC, "thread_exit: thread %p (pid %ld, %s)", td, 942 (long)p->p_pid, td->td_name); 943 SDT_PROBE0(proc, , , lwp__exit); 944 KASSERT(TAILQ_EMPTY(&td->td_sigqueue.sq_list), ("signal pending")); 945 MPASS(td->td_realucred == td->td_ucred); 946 947 /* 948 * drop FPU & debug register state storage, or any other 949 * architecture specific resources that 950 * would not be on a new untouched process. 951 */ 952 cpu_thread_exit(td); 953 954 /* 955 * The last thread is left attached to the process 956 * So that the whole bundle gets recycled. Skip 957 * all this stuff if we never had threads. 958 * EXIT clears all sign of other threads when 959 * it goes to single threading, so the last thread always 960 * takes the short path. 961 */ 962 if (p->p_flag & P_HADTHREADS) { 963 if (p->p_numthreads > 1) { 964 atomic_add_int(&td->td_proc->p_exitthreads, 1); 965 thread_unlink(td); 966 td2 = FIRST_THREAD_IN_PROC(p); 967 sched_exit_thread(td2, td); 968 969 /* 970 * The test below is NOT true if we are the 971 * sole exiting thread. P_STOPPED_SINGLE is unset 972 * in exit1() after it is the only survivor. 973 */ 974 if (P_SHOULDSTOP(p) == P_STOPPED_SINGLE) { 975 if (p->p_numthreads == p->p_suspcount) { 976 thread_lock(p->p_singlethread); 977 wakeup_swapper = thread_unsuspend_one( 978 p->p_singlethread, p, false); 979 if (wakeup_swapper) 980 kick_proc0(); 981 } 982 } 983 984 PCPU_SET(deadthread, td); 985 } else { 986 /* 987 * The last thread is exiting.. but not through exit() 988 */ 989 panic ("thread_exit: Last thread exiting on its own"); 990 } 991 } 992 #ifdef HWPMC_HOOKS 993 /* 994 * If this thread is part of a process that is being tracked by hwpmc(4), 995 * inform the module of the thread's impending exit. 996 */ 997 if (PMC_PROC_IS_USING_PMCS(td->td_proc)) { 998 PMC_SWITCH_CONTEXT(td, PMC_FN_CSW_OUT); 999 PMC_CALL_HOOK_UNLOCKED(td, PMC_FN_THR_EXIT, NULL); 1000 } else if (PMC_SYSTEM_SAMPLING_ACTIVE()) 1001 PMC_CALL_HOOK_UNLOCKED(td, PMC_FN_THR_EXIT_LOG, NULL); 1002 #endif 1003 PROC_UNLOCK(p); 1004 PROC_STATLOCK(p); 1005 thread_lock(td); 1006 PROC_SUNLOCK(p); 1007 1008 /* Do the same timestamp bookkeeping that mi_switch() would do. */ 1009 new_switchtime = cpu_ticks(); 1010 runtime = new_switchtime - PCPU_GET(switchtime); 1011 td->td_runtime += runtime; 1012 td->td_incruntime += runtime; 1013 PCPU_SET(switchtime, new_switchtime); 1014 PCPU_SET(switchticks, ticks); 1015 VM_CNT_INC(v_swtch); 1016 1017 /* Save our resource usage in our process. */ 1018 td->td_ru.ru_nvcsw++; 1019 ruxagg_locked(p, td); 1020 rucollect(&p->p_ru, &td->td_ru); 1021 PROC_STATUNLOCK(p); 1022 1023 TD_SET_STATE(td, TDS_INACTIVE); 1024 #ifdef WITNESS 1025 witness_thread_exit(td); 1026 #endif 1027 CTR1(KTR_PROC, "thread_exit: cpu_throw() thread %p", td); 1028 sched_throw(td); 1029 panic("I'm a teapot!"); 1030 /* NOTREACHED */ 1031 } 1032 1033 /* 1034 * Do any thread specific cleanups that may be needed in wait() 1035 * called with Giant, proc and schedlock not held. 1036 */ 1037 void 1038 thread_wait(struct proc *p) 1039 { 1040 struct thread *td; 1041 1042 mtx_assert(&Giant, MA_NOTOWNED); 1043 KASSERT(p->p_numthreads == 1, ("multiple threads in thread_wait()")); 1044 KASSERT(p->p_exitthreads == 0, ("p_exitthreads leaking")); 1045 td = FIRST_THREAD_IN_PROC(p); 1046 /* Lock the last thread so we spin until it exits cpu_throw(). */ 1047 thread_lock(td); 1048 thread_unlock(td); 1049 lock_profile_thread_exit(td); 1050 cpuset_rel(td->td_cpuset); 1051 td->td_cpuset = NULL; 1052 cpu_thread_clean(td); 1053 thread_cow_free(td); 1054 callout_drain(&td->td_slpcallout); 1055 thread_reap(); /* check for zombie threads etc. */ 1056 } 1057 1058 /* 1059 * Link a thread to a process. 1060 * set up anything that needs to be initialized for it to 1061 * be used by the process. 1062 */ 1063 void 1064 thread_link(struct thread *td, struct proc *p) 1065 { 1066 1067 /* 1068 * XXX This can't be enabled because it's called for proc0 before 1069 * its lock has been created. 1070 * PROC_LOCK_ASSERT(p, MA_OWNED); 1071 */ 1072 TD_SET_STATE(td, TDS_INACTIVE); 1073 td->td_proc = p; 1074 td->td_flags = TDF_INMEM; 1075 1076 LIST_INIT(&td->td_contested); 1077 LIST_INIT(&td->td_lprof[0]); 1078 LIST_INIT(&td->td_lprof[1]); 1079 #ifdef EPOCH_TRACE 1080 SLIST_INIT(&td->td_epochs); 1081 #endif 1082 sigqueue_init(&td->td_sigqueue, p); 1083 callout_init(&td->td_slpcallout, 1); 1084 TAILQ_INSERT_TAIL(&p->p_threads, td, td_plist); 1085 p->p_numthreads++; 1086 } 1087 1088 /* 1089 * Called from: 1090 * thread_exit() 1091 */ 1092 void 1093 thread_unlink(struct thread *td) 1094 { 1095 struct proc *p = td->td_proc; 1096 1097 PROC_LOCK_ASSERT(p, MA_OWNED); 1098 #ifdef EPOCH_TRACE 1099 MPASS(SLIST_EMPTY(&td->td_epochs)); 1100 #endif 1101 1102 TAILQ_REMOVE(&p->p_threads, td, td_plist); 1103 p->p_numthreads--; 1104 /* could clear a few other things here */ 1105 /* Must NOT clear links to proc! */ 1106 } 1107 1108 static int 1109 calc_remaining(struct proc *p, int mode) 1110 { 1111 int remaining; 1112 1113 PROC_LOCK_ASSERT(p, MA_OWNED); 1114 PROC_SLOCK_ASSERT(p, MA_OWNED); 1115 if (mode == SINGLE_EXIT) 1116 remaining = p->p_numthreads; 1117 else if (mode == SINGLE_BOUNDARY) 1118 remaining = p->p_numthreads - p->p_boundary_count; 1119 else if (mode == SINGLE_NO_EXIT || mode == SINGLE_ALLPROC) 1120 remaining = p->p_numthreads - p->p_suspcount; 1121 else 1122 panic("calc_remaining: wrong mode %d", mode); 1123 return (remaining); 1124 } 1125 1126 static int 1127 remain_for_mode(int mode) 1128 { 1129 1130 return (mode == SINGLE_ALLPROC ? 0 : 1); 1131 } 1132 1133 static int 1134 weed_inhib(int mode, struct thread *td2, struct proc *p) 1135 { 1136 int wakeup_swapper; 1137 1138 PROC_LOCK_ASSERT(p, MA_OWNED); 1139 PROC_SLOCK_ASSERT(p, MA_OWNED); 1140 THREAD_LOCK_ASSERT(td2, MA_OWNED); 1141 1142 wakeup_swapper = 0; 1143 1144 /* 1145 * Since the thread lock is dropped by the scheduler we have 1146 * to retry to check for races. 1147 */ 1148 restart: 1149 switch (mode) { 1150 case SINGLE_EXIT: 1151 if (TD_IS_SUSPENDED(td2)) { 1152 wakeup_swapper |= thread_unsuspend_one(td2, p, true); 1153 thread_lock(td2); 1154 goto restart; 1155 } 1156 if (TD_CAN_ABORT(td2)) { 1157 wakeup_swapper |= sleepq_abort(td2, EINTR); 1158 return (wakeup_swapper); 1159 } 1160 break; 1161 case SINGLE_BOUNDARY: 1162 case SINGLE_NO_EXIT: 1163 if (TD_IS_SUSPENDED(td2) && 1164 (td2->td_flags & TDF_BOUNDARY) == 0) { 1165 wakeup_swapper |= thread_unsuspend_one(td2, p, false); 1166 thread_lock(td2); 1167 goto restart; 1168 } 1169 if (TD_CAN_ABORT(td2)) { 1170 wakeup_swapper |= sleepq_abort(td2, ERESTART); 1171 return (wakeup_swapper); 1172 } 1173 break; 1174 case SINGLE_ALLPROC: 1175 /* 1176 * ALLPROC suspend tries to avoid spurious EINTR for 1177 * threads sleeping interruptable, by suspending the 1178 * thread directly, similarly to sig_suspend_threads(). 1179 * Since such sleep is not neccessary performed at the user 1180 * boundary, TDF_ALLPROCSUSP is used to avoid immediate 1181 * un-suspend. 1182 */ 1183 if (TD_IS_SUSPENDED(td2) && 1184 (td2->td_flags & TDF_ALLPROCSUSP) == 0) { 1185 wakeup_swapper |= thread_unsuspend_one(td2, p, false); 1186 thread_lock(td2); 1187 goto restart; 1188 } 1189 if (TD_CAN_ABORT(td2)) { 1190 td2->td_flags |= TDF_ALLPROCSUSP; 1191 wakeup_swapper |= sleepq_abort(td2, ERESTART); 1192 return (wakeup_swapper); 1193 } 1194 break; 1195 default: 1196 break; 1197 } 1198 thread_unlock(td2); 1199 return (wakeup_swapper); 1200 } 1201 1202 /* 1203 * Enforce single-threading. 1204 * 1205 * Returns 1 if the caller must abort (another thread is waiting to 1206 * exit the process or similar). Process is locked! 1207 * Returns 0 when you are successfully the only thread running. 1208 * A process has successfully single threaded in the suspend mode when 1209 * There are no threads in user mode. Threads in the kernel must be 1210 * allowed to continue until they get to the user boundary. They may even 1211 * copy out their return values and data before suspending. They may however be 1212 * accelerated in reaching the user boundary as we will wake up 1213 * any sleeping threads that are interruptable. (PCATCH). 1214 */ 1215 int 1216 thread_single(struct proc *p, int mode) 1217 { 1218 struct thread *td; 1219 struct thread *td2; 1220 int remaining, wakeup_swapper; 1221 1222 td = curthread; 1223 KASSERT(mode == SINGLE_EXIT || mode == SINGLE_BOUNDARY || 1224 mode == SINGLE_ALLPROC || mode == SINGLE_NO_EXIT, 1225 ("invalid mode %d", mode)); 1226 /* 1227 * If allowing non-ALLPROC singlethreading for non-curproc 1228 * callers, calc_remaining() and remain_for_mode() should be 1229 * adjusted to also account for td->td_proc != p. For now 1230 * this is not implemented because it is not used. 1231 */ 1232 KASSERT((mode == SINGLE_ALLPROC && td->td_proc != p) || 1233 (mode != SINGLE_ALLPROC && td->td_proc == p), 1234 ("mode %d proc %p curproc %p", mode, p, td->td_proc)); 1235 mtx_assert(&Giant, MA_NOTOWNED); 1236 PROC_LOCK_ASSERT(p, MA_OWNED); 1237 1238 /* 1239 * Is someone already single threading? 1240 * Or may be singlethreading is not needed at all. 1241 */ 1242 if (mode == SINGLE_ALLPROC) { 1243 while ((p->p_flag & P_STOPPED_SINGLE) != 0) { 1244 if ((p->p_flag2 & P2_WEXIT) != 0) 1245 return (1); 1246 msleep(&p->p_flag, &p->p_mtx, PCATCH, "thrsgl", 0); 1247 } 1248 } else if ((p->p_flag & P_HADTHREADS) == 0) 1249 return (0); 1250 if (p->p_singlethread != NULL && p->p_singlethread != td) 1251 return (1); 1252 1253 if (mode == SINGLE_EXIT) { 1254 p->p_flag |= P_SINGLE_EXIT; 1255 p->p_flag &= ~P_SINGLE_BOUNDARY; 1256 } else { 1257 p->p_flag &= ~P_SINGLE_EXIT; 1258 if (mode == SINGLE_BOUNDARY) 1259 p->p_flag |= P_SINGLE_BOUNDARY; 1260 else 1261 p->p_flag &= ~P_SINGLE_BOUNDARY; 1262 } 1263 if (mode == SINGLE_ALLPROC) 1264 p->p_flag |= P_TOTAL_STOP; 1265 p->p_flag |= P_STOPPED_SINGLE; 1266 PROC_SLOCK(p); 1267 p->p_singlethread = td; 1268 remaining = calc_remaining(p, mode); 1269 while (remaining != remain_for_mode(mode)) { 1270 if (P_SHOULDSTOP(p) != P_STOPPED_SINGLE) 1271 goto stopme; 1272 wakeup_swapper = 0; 1273 FOREACH_THREAD_IN_PROC(p, td2) { 1274 if (td2 == td) 1275 continue; 1276 thread_lock(td2); 1277 ast_sched_locked(td2, TDA_SUSPEND); 1278 if (TD_IS_INHIBITED(td2)) { 1279 wakeup_swapper |= weed_inhib(mode, td2, p); 1280 #ifdef SMP 1281 } else if (TD_IS_RUNNING(td2)) { 1282 forward_signal(td2); 1283 thread_unlock(td2); 1284 #endif 1285 } else 1286 thread_unlock(td2); 1287 } 1288 if (wakeup_swapper) 1289 kick_proc0(); 1290 remaining = calc_remaining(p, mode); 1291 1292 /* 1293 * Maybe we suspended some threads.. was it enough? 1294 */ 1295 if (remaining == remain_for_mode(mode)) 1296 break; 1297 1298 stopme: 1299 /* 1300 * Wake us up when everyone else has suspended. 1301 * In the mean time we suspend as well. 1302 */ 1303 thread_suspend_switch(td, p); 1304 remaining = calc_remaining(p, mode); 1305 } 1306 if (mode == SINGLE_EXIT) { 1307 /* 1308 * Convert the process to an unthreaded process. The 1309 * SINGLE_EXIT is called by exit1() or execve(), in 1310 * both cases other threads must be retired. 1311 */ 1312 KASSERT(p->p_numthreads == 1, ("Unthreading with >1 threads")); 1313 p->p_singlethread = NULL; 1314 p->p_flag &= ~(P_STOPPED_SINGLE | P_SINGLE_EXIT | P_HADTHREADS); 1315 1316 /* 1317 * Wait for any remaining threads to exit cpu_throw(). 1318 */ 1319 while (p->p_exitthreads != 0) { 1320 PROC_SUNLOCK(p); 1321 PROC_UNLOCK(p); 1322 sched_relinquish(td); 1323 PROC_LOCK(p); 1324 PROC_SLOCK(p); 1325 } 1326 } else if (mode == SINGLE_BOUNDARY) { 1327 /* 1328 * Wait until all suspended threads are removed from 1329 * the processors. The thread_suspend_check() 1330 * increments p_boundary_count while it is still 1331 * running, which makes it possible for the execve() 1332 * to destroy vmspace while our other threads are 1333 * still using the address space. 1334 * 1335 * We lock the thread, which is only allowed to 1336 * succeed after context switch code finished using 1337 * the address space. 1338 */ 1339 FOREACH_THREAD_IN_PROC(p, td2) { 1340 if (td2 == td) 1341 continue; 1342 thread_lock(td2); 1343 KASSERT((td2->td_flags & TDF_BOUNDARY) != 0, 1344 ("td %p not on boundary", td2)); 1345 KASSERT(TD_IS_SUSPENDED(td2), 1346 ("td %p is not suspended", td2)); 1347 thread_unlock(td2); 1348 } 1349 } 1350 PROC_SUNLOCK(p); 1351 return (0); 1352 } 1353 1354 bool 1355 thread_suspend_check_needed(void) 1356 { 1357 struct proc *p; 1358 struct thread *td; 1359 1360 td = curthread; 1361 p = td->td_proc; 1362 PROC_LOCK_ASSERT(p, MA_OWNED); 1363 return (P_SHOULDSTOP(p) || ((p->p_flag & P_TRACED) != 0 && 1364 (td->td_dbgflags & TDB_SUSPEND) != 0)); 1365 } 1366 1367 /* 1368 * Called in from locations that can safely check to see 1369 * whether we have to suspend or at least throttle for a 1370 * single-thread event (e.g. fork). 1371 * 1372 * Such locations include userret(). 1373 * If the "return_instead" argument is non zero, the thread must be able to 1374 * accept 0 (caller may continue), or 1 (caller must abort) as a result. 1375 * 1376 * The 'return_instead' argument tells the function if it may do a 1377 * thread_exit() or suspend, or whether the caller must abort and back 1378 * out instead. 1379 * 1380 * If the thread that set the single_threading request has set the 1381 * P_SINGLE_EXIT bit in the process flags then this call will never return 1382 * if 'return_instead' is false, but will exit. 1383 * 1384 * P_SINGLE_EXIT | return_instead == 0| return_instead != 0 1385 *---------------+--------------------+--------------------- 1386 * 0 | returns 0 | returns 0 or 1 1387 * | when ST ends | immediately 1388 *---------------+--------------------+--------------------- 1389 * 1 | thread exits | returns 1 1390 * | | immediately 1391 * 0 = thread_exit() or suspension ok, 1392 * other = return error instead of stopping the thread. 1393 * 1394 * While a full suspension is under effect, even a single threading 1395 * thread would be suspended if it made this call (but it shouldn't). 1396 * This call should only be made from places where 1397 * thread_exit() would be safe as that may be the outcome unless 1398 * return_instead is set. 1399 */ 1400 int 1401 thread_suspend_check(int return_instead) 1402 { 1403 struct thread *td; 1404 struct proc *p; 1405 int wakeup_swapper; 1406 1407 td = curthread; 1408 p = td->td_proc; 1409 mtx_assert(&Giant, MA_NOTOWNED); 1410 PROC_LOCK_ASSERT(p, MA_OWNED); 1411 while (thread_suspend_check_needed()) { 1412 if (P_SHOULDSTOP(p) == P_STOPPED_SINGLE) { 1413 KASSERT(p->p_singlethread != NULL, 1414 ("singlethread not set")); 1415 /* 1416 * The only suspension in action is a 1417 * single-threading. Single threader need not stop. 1418 * It is safe to access p->p_singlethread unlocked 1419 * because it can only be set to our address by us. 1420 */ 1421 if (p->p_singlethread == td) 1422 return (0); /* Exempt from stopping. */ 1423 } 1424 if ((p->p_flag & P_SINGLE_EXIT) && return_instead) 1425 return (EINTR); 1426 1427 /* Should we goto user boundary if we didn't come from there? */ 1428 if (P_SHOULDSTOP(p) == P_STOPPED_SINGLE && 1429 (p->p_flag & P_SINGLE_BOUNDARY) && return_instead) 1430 return (ERESTART); 1431 1432 /* 1433 * Ignore suspend requests if they are deferred. 1434 */ 1435 if ((td->td_flags & TDF_SBDRY) != 0) { 1436 KASSERT(return_instead, 1437 ("TDF_SBDRY set for unsafe thread_suspend_check")); 1438 KASSERT((td->td_flags & (TDF_SEINTR | TDF_SERESTART)) != 1439 (TDF_SEINTR | TDF_SERESTART), 1440 ("both TDF_SEINTR and TDF_SERESTART")); 1441 return (TD_SBDRY_INTR(td) ? TD_SBDRY_ERRNO(td) : 0); 1442 } 1443 1444 /* 1445 * If the process is waiting for us to exit, 1446 * this thread should just suicide. 1447 * Assumes that P_SINGLE_EXIT implies P_STOPPED_SINGLE. 1448 */ 1449 if ((p->p_flag & P_SINGLE_EXIT) && (p->p_singlethread != td)) { 1450 PROC_UNLOCK(p); 1451 1452 /* 1453 * Allow Linux emulation layer to do some work 1454 * before thread suicide. 1455 */ 1456 if (__predict_false(p->p_sysent->sv_thread_detach != NULL)) 1457 (p->p_sysent->sv_thread_detach)(td); 1458 umtx_thread_exit(td); 1459 kern_thr_exit(td); 1460 panic("stopped thread did not exit"); 1461 } 1462 1463 PROC_SLOCK(p); 1464 thread_stopped(p); 1465 if (P_SHOULDSTOP(p) == P_STOPPED_SINGLE) { 1466 if (p->p_numthreads == p->p_suspcount + 1) { 1467 thread_lock(p->p_singlethread); 1468 wakeup_swapper = thread_unsuspend_one( 1469 p->p_singlethread, p, false); 1470 if (wakeup_swapper) 1471 kick_proc0(); 1472 } 1473 } 1474 PROC_UNLOCK(p); 1475 thread_lock(td); 1476 /* 1477 * When a thread suspends, it just 1478 * gets taken off all queues. 1479 */ 1480 thread_suspend_one(td); 1481 if (return_instead == 0) { 1482 p->p_boundary_count++; 1483 td->td_flags |= TDF_BOUNDARY; 1484 } 1485 PROC_SUNLOCK(p); 1486 mi_switch(SW_INVOL | SWT_SUSPEND); 1487 PROC_LOCK(p); 1488 } 1489 return (0); 1490 } 1491 1492 /* 1493 * Check for possible stops and suspensions while executing a 1494 * casueword or similar transiently failing operation. 1495 * 1496 * The sleep argument controls whether the function can handle a stop 1497 * request itself or it should return ERESTART and the request is 1498 * proceed at the kernel/user boundary in ast. 1499 * 1500 * Typically, when retrying due to casueword(9) failure (rv == 1), we 1501 * should handle the stop requests there, with exception of cases when 1502 * the thread owns a kernel resource, for instance busied the umtx 1503 * key, or when functions return immediately if thread_check_susp() 1504 * returned non-zero. On the other hand, retrying the whole lock 1505 * operation, we better not stop there but delegate the handling to 1506 * ast. 1507 * 1508 * If the request is for thread termination P_SINGLE_EXIT, we cannot 1509 * handle it at all, and simply return EINTR. 1510 */ 1511 int 1512 thread_check_susp(struct thread *td, bool sleep) 1513 { 1514 struct proc *p; 1515 int error; 1516 1517 /* 1518 * The check for TDA_SUSPEND is racy, but it is enough to 1519 * eventually break the lockstep loop. 1520 */ 1521 if (!td_ast_pending(td, TDA_SUSPEND)) 1522 return (0); 1523 error = 0; 1524 p = td->td_proc; 1525 PROC_LOCK(p); 1526 if (p->p_flag & P_SINGLE_EXIT) 1527 error = EINTR; 1528 else if (P_SHOULDSTOP(p) || 1529 ((p->p_flag & P_TRACED) && (td->td_dbgflags & TDB_SUSPEND))) 1530 error = sleep ? thread_suspend_check(0) : ERESTART; 1531 PROC_UNLOCK(p); 1532 return (error); 1533 } 1534 1535 void 1536 thread_suspend_switch(struct thread *td, struct proc *p) 1537 { 1538 1539 KASSERT(!TD_IS_SUSPENDED(td), ("already suspended")); 1540 PROC_LOCK_ASSERT(p, MA_OWNED); 1541 PROC_SLOCK_ASSERT(p, MA_OWNED); 1542 /* 1543 * We implement thread_suspend_one in stages here to avoid 1544 * dropping the proc lock while the thread lock is owned. 1545 */ 1546 if (p == td->td_proc) { 1547 thread_stopped(p); 1548 p->p_suspcount++; 1549 } 1550 PROC_UNLOCK(p); 1551 thread_lock(td); 1552 ast_unsched_locked(td, TDA_SUSPEND); 1553 TD_SET_SUSPENDED(td); 1554 sched_sleep(td, 0); 1555 PROC_SUNLOCK(p); 1556 DROP_GIANT(); 1557 mi_switch(SW_VOL | SWT_SUSPEND); 1558 PICKUP_GIANT(); 1559 PROC_LOCK(p); 1560 PROC_SLOCK(p); 1561 } 1562 1563 void 1564 thread_suspend_one(struct thread *td) 1565 { 1566 struct proc *p; 1567 1568 p = td->td_proc; 1569 PROC_SLOCK_ASSERT(p, MA_OWNED); 1570 THREAD_LOCK_ASSERT(td, MA_OWNED); 1571 KASSERT(!TD_IS_SUSPENDED(td), ("already suspended")); 1572 p->p_suspcount++; 1573 ast_unsched_locked(td, TDA_SUSPEND); 1574 TD_SET_SUSPENDED(td); 1575 sched_sleep(td, 0); 1576 } 1577 1578 static int 1579 thread_unsuspend_one(struct thread *td, struct proc *p, bool boundary) 1580 { 1581 1582 THREAD_LOCK_ASSERT(td, MA_OWNED); 1583 KASSERT(TD_IS_SUSPENDED(td), ("Thread not suspended")); 1584 TD_CLR_SUSPENDED(td); 1585 td->td_flags &= ~TDF_ALLPROCSUSP; 1586 if (td->td_proc == p) { 1587 PROC_SLOCK_ASSERT(p, MA_OWNED); 1588 p->p_suspcount--; 1589 if (boundary && (td->td_flags & TDF_BOUNDARY) != 0) { 1590 td->td_flags &= ~TDF_BOUNDARY; 1591 p->p_boundary_count--; 1592 } 1593 } 1594 return (setrunnable(td, 0)); 1595 } 1596 1597 void 1598 thread_run_flash(struct thread *td) 1599 { 1600 struct proc *p; 1601 1602 p = td->td_proc; 1603 PROC_LOCK_ASSERT(p, MA_OWNED); 1604 1605 if (TD_ON_SLEEPQ(td)) 1606 sleepq_remove_nested(td); 1607 else 1608 thread_lock(td); 1609 1610 THREAD_LOCK_ASSERT(td, MA_OWNED); 1611 KASSERT(TD_IS_SUSPENDED(td), ("Thread not suspended")); 1612 1613 TD_CLR_SUSPENDED(td); 1614 PROC_SLOCK(p); 1615 MPASS(p->p_suspcount > 0); 1616 p->p_suspcount--; 1617 PROC_SUNLOCK(p); 1618 if (setrunnable(td, 0)) 1619 kick_proc0(); 1620 } 1621 1622 /* 1623 * Allow all threads blocked by single threading to continue running. 1624 */ 1625 void 1626 thread_unsuspend(struct proc *p) 1627 { 1628 struct thread *td; 1629 int wakeup_swapper; 1630 1631 PROC_LOCK_ASSERT(p, MA_OWNED); 1632 PROC_SLOCK_ASSERT(p, MA_OWNED); 1633 wakeup_swapper = 0; 1634 if (!P_SHOULDSTOP(p)) { 1635 FOREACH_THREAD_IN_PROC(p, td) { 1636 thread_lock(td); 1637 if (TD_IS_SUSPENDED(td)) 1638 wakeup_swapper |= thread_unsuspend_one(td, p, 1639 true); 1640 else 1641 thread_unlock(td); 1642 } 1643 } else if (P_SHOULDSTOP(p) == P_STOPPED_SINGLE && 1644 p->p_numthreads == p->p_suspcount) { 1645 /* 1646 * Stopping everything also did the job for the single 1647 * threading request. Now we've downgraded to single-threaded, 1648 * let it continue. 1649 */ 1650 if (p->p_singlethread->td_proc == p) { 1651 thread_lock(p->p_singlethread); 1652 wakeup_swapper = thread_unsuspend_one( 1653 p->p_singlethread, p, false); 1654 } 1655 } 1656 if (wakeup_swapper) 1657 kick_proc0(); 1658 } 1659 1660 /* 1661 * End the single threading mode.. 1662 */ 1663 void 1664 thread_single_end(struct proc *p, int mode) 1665 { 1666 struct thread *td; 1667 int wakeup_swapper; 1668 1669 KASSERT(mode == SINGLE_EXIT || mode == SINGLE_BOUNDARY || 1670 mode == SINGLE_ALLPROC || mode == SINGLE_NO_EXIT, 1671 ("invalid mode %d", mode)); 1672 PROC_LOCK_ASSERT(p, MA_OWNED); 1673 KASSERT((mode == SINGLE_ALLPROC && (p->p_flag & P_TOTAL_STOP) != 0) || 1674 (mode != SINGLE_ALLPROC && (p->p_flag & P_TOTAL_STOP) == 0), 1675 ("mode %d does not match P_TOTAL_STOP", mode)); 1676 KASSERT(mode == SINGLE_ALLPROC || p->p_singlethread == curthread, 1677 ("thread_single_end from other thread %p %p", 1678 curthread, p->p_singlethread)); 1679 KASSERT(mode != SINGLE_BOUNDARY || 1680 (p->p_flag & P_SINGLE_BOUNDARY) != 0, 1681 ("mis-matched SINGLE_BOUNDARY flags %x", p->p_flag)); 1682 p->p_flag &= ~(P_STOPPED_SINGLE | P_SINGLE_EXIT | P_SINGLE_BOUNDARY | 1683 P_TOTAL_STOP); 1684 PROC_SLOCK(p); 1685 p->p_singlethread = NULL; 1686 wakeup_swapper = 0; 1687 /* 1688 * If there are other threads they may now run, 1689 * unless of course there is a blanket 'stop order' 1690 * on the process. The single threader must be allowed 1691 * to continue however as this is a bad place to stop. 1692 */ 1693 if (p->p_numthreads != remain_for_mode(mode) && !P_SHOULDSTOP(p)) { 1694 FOREACH_THREAD_IN_PROC(p, td) { 1695 thread_lock(td); 1696 if (TD_IS_SUSPENDED(td)) { 1697 wakeup_swapper |= thread_unsuspend_one(td, p, 1698 true); 1699 } else 1700 thread_unlock(td); 1701 } 1702 } 1703 KASSERT(mode != SINGLE_BOUNDARY || p->p_boundary_count == 0, 1704 ("inconsistent boundary count %d", p->p_boundary_count)); 1705 PROC_SUNLOCK(p); 1706 if (wakeup_swapper) 1707 kick_proc0(); 1708 wakeup(&p->p_flag); 1709 } 1710 1711 /* 1712 * Locate a thread by number and return with proc lock held. 1713 * 1714 * thread exit establishes proc -> tidhash lock ordering, but lookup 1715 * takes tidhash first and needs to return locked proc. 1716 * 1717 * The problem is worked around by relying on type-safety of both 1718 * structures and doing the work in 2 steps: 1719 * - tidhash-locked lookup which saves both thread and proc pointers 1720 * - proc-locked verification that the found thread still matches 1721 */ 1722 static bool 1723 tdfind_hash(lwpid_t tid, pid_t pid, struct proc **pp, struct thread **tdp) 1724 { 1725 #define RUN_THRESH 16 1726 struct proc *p; 1727 struct thread *td; 1728 int run; 1729 bool locked; 1730 1731 run = 0; 1732 rw_rlock(TIDHASHLOCK(tid)); 1733 locked = true; 1734 LIST_FOREACH(td, TIDHASH(tid), td_hash) { 1735 if (td->td_tid != tid) { 1736 run++; 1737 continue; 1738 } 1739 p = td->td_proc; 1740 if (pid != -1 && p->p_pid != pid) { 1741 td = NULL; 1742 break; 1743 } 1744 if (run > RUN_THRESH) { 1745 if (rw_try_upgrade(TIDHASHLOCK(tid))) { 1746 LIST_REMOVE(td, td_hash); 1747 LIST_INSERT_HEAD(TIDHASH(td->td_tid), 1748 td, td_hash); 1749 rw_wunlock(TIDHASHLOCK(tid)); 1750 locked = false; 1751 break; 1752 } 1753 } 1754 break; 1755 } 1756 if (locked) 1757 rw_runlock(TIDHASHLOCK(tid)); 1758 if (td == NULL) 1759 return (false); 1760 *pp = p; 1761 *tdp = td; 1762 return (true); 1763 } 1764 1765 struct thread * 1766 tdfind(lwpid_t tid, pid_t pid) 1767 { 1768 struct proc *p; 1769 struct thread *td; 1770 1771 td = curthread; 1772 if (td->td_tid == tid) { 1773 if (pid != -1 && td->td_proc->p_pid != pid) 1774 return (NULL); 1775 PROC_LOCK(td->td_proc); 1776 return (td); 1777 } 1778 1779 for (;;) { 1780 if (!tdfind_hash(tid, pid, &p, &td)) 1781 return (NULL); 1782 PROC_LOCK(p); 1783 if (td->td_tid != tid) { 1784 PROC_UNLOCK(p); 1785 continue; 1786 } 1787 if (td->td_proc != p) { 1788 PROC_UNLOCK(p); 1789 continue; 1790 } 1791 if (p->p_state == PRS_NEW) { 1792 PROC_UNLOCK(p); 1793 return (NULL); 1794 } 1795 return (td); 1796 } 1797 } 1798 1799 void 1800 tidhash_add(struct thread *td) 1801 { 1802 rw_wlock(TIDHASHLOCK(td->td_tid)); 1803 LIST_INSERT_HEAD(TIDHASH(td->td_tid), td, td_hash); 1804 rw_wunlock(TIDHASHLOCK(td->td_tid)); 1805 } 1806 1807 void 1808 tidhash_remove(struct thread *td) 1809 { 1810 1811 rw_wlock(TIDHASHLOCK(td->td_tid)); 1812 LIST_REMOVE(td, td_hash); 1813 rw_wunlock(TIDHASHLOCK(td->td_tid)); 1814 } 1815