sys/kern/kern_thread.c

   1 /*-
   2  * Copyright (C) 2001 Julian Elischer <julian@freebsd.org>.
   3  *  All rights reserved.
   4  *
   5  * Redistribution and use in source and binary forms, with or without
   6  * modification, are permitted provided that the following conditions
   7  * are met:
   8  * 1. Redistributions of source code must retain the above copyright
   9  *    notice(s), this list of conditions and the following disclaimer as
  10  *    the first lines of this file unmodified other than the possible
  11  *    addition of one or more copyright notices.
  12  * 2. Redistributions in binary form must reproduce the above copyright
  13  *    notice(s), this list of conditions and the following disclaimer in the
  14  *    documentation and/or other materials provided with the distribution.
  15  *
  16  * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDER(S) ``AS IS'' AND ANY
  17  * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
  18  * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
  19  * DISCLAIMED.  IN NO EVENT SHALL THE COPYRIGHT HOLDER(S) BE LIABLE FOR ANY
  20  * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
  21  * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
  22  * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
  23  * CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
  24  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
  25  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH
  26  * DAMAGE.
  27  */
  28
  29 #include "opt_witness.h"
  30 #include "opt_hwpmc_hooks.h"
  31
  32 #include <sys/cdefs.h>
  33 __FBSDID("$FreeBSD$");
  34
  35 #include <sys/param.h>
  36 #include <sys/systm.h>
  37 #include <sys/kernel.h>
  38 #include <sys/lock.h>
  39 #include <sys/mutex.h>
  40 #include <sys/proc.h>
  41 #include <sys/rangelock.h>
  42 #include <sys/resourcevar.h>
  43 #include <sys/sdt.h>
  44 #include <sys/smp.h>
  45 #include <sys/sched.h>
  46 #include <sys/sleepqueue.h>
  47 #include <sys/selinfo.h>
  48 #include <sys/turnstile.h>
  49 #include <sys/ktr.h>
  50 #include <sys/rwlock.h>
  51 #include <sys/umtx.h>
  52 #include <sys/cpuset.h>
  53 #ifdef  HWPMC_HOOKS
  54 #include <sys/pmckern.h>
  55 #endif
  56
  57 #include <security/audit/audit.h>
  58
  59 #include <vm/vm.h>
  60 #include <vm/vm_extern.h>
  61 #include <vm/uma.h>
  62 #include <sys/eventhandler.h>
  63
  64 SDT_PROVIDER_DECLARE(proc);
  65 SDT_PROBE_DEFINE(proc, , , lwp__exit);
  66
  67
  68 /*
  69  * thread related storage.
  70  */
  71 static uma_zone_t thread_zone;
  72
  73 TAILQ_HEAD(, thread) zombie_threads = TAILQ_HEAD_INITIALIZER(zombie_threads);
  74 static struct mtx zombie_lock;
  75 MTX_SYSINIT(zombie_lock, &zombie_lock, "zombie lock", MTX_SPIN);
  76
  77 static void thread_zombie(struct thread *);
  78
  79 #define TID_BUFFER_SIZE 1024
  80
  81 struct mtx tid_lock;
  82 static struct unrhdr *tid_unrhdr;
  83 static lwpid_t tid_buffer[TID_BUFFER_SIZE];
  84 static int tid_head, tid_tail;
  85 static MALLOC_DEFINE(M_TIDHASH, "tidhash", "thread hash");
  86
  87 struct  tidhashhead *tidhashtbl;
  88 u_long  tidhash;
  89 struct  rwlock tidhash_lock;
  90
  91 static lwpid_t
  92 tid_alloc(void)
  93 {
  94         lwpid_t tid;
  95
  96         tid = alloc_unr(tid_unrhdr);
  97         if (tid != -1)
  98                 return (tid);
  99         mtx_lock(&tid_lock);
 100         if (tid_head == tid_tail) {
 101                 mtx_unlock(&tid_lock);
 102                 return (-1);
 103         }
 104         tid = tid_buffer[tid_head];
 105         tid_head = (tid_head + 1) % TID_BUFFER_SIZE;
 106         mtx_unlock(&tid_lock);
 107         return (tid);
 108 }
 109
 110 static void
 111 tid_free(lwpid_t tid)
 112 {
 113         lwpid_t tmp_tid = -1;
 114
 115         mtx_lock(&tid_lock);
 116         if ((tid_tail + 1) % TID_BUFFER_SIZE == tid_head) {
 117                 tmp_tid = tid_buffer[tid_head];
 118                 tid_head = (tid_head + 1) % TID_BUFFER_SIZE;
 119         }
 120         tid_buffer[tid_tail] = tid;
 121         tid_tail = (tid_tail + 1) % TID_BUFFER_SIZE;
 122         mtx_unlock(&tid_lock);
 123         if (tmp_tid != -1)
 124                 free_unr(tid_unrhdr, tmp_tid);
 125 }
 126
 127 /*
 128  * Prepare a thread for use.
 129  */
 130 static int
 131 thread_ctor(void *mem, int size, void *arg, int flags)
 132 {
 133         struct thread   *td;
 134
 135         td = (struct thread *)mem;
 136         td->td_state = TDS_INACTIVE;
 137         td->td_oncpu = NOCPU;
 138
 139         td->td_tid = tid_alloc();
 140
 141         /*
 142          * Note that td_critnest begins life as 1 because the thread is not
 143          * running and is thereby implicitly waiting to be on the receiving
 144          * end of a context switch.
 145          */
 146         td->td_critnest = 1;
 147         td->td_lend_user_pri = PRI_MAX;
 148         EVENTHANDLER_INVOKE(thread_ctor, td);
 149 #ifdef AUDIT
 150         audit_thread_alloc(td);
 151 #endif
 152         umtx_thread_alloc(td);
 153         return (0);
 154 }
 155
 156 /*
 157  * Reclaim a thread after use.
 158  */
 159 static void
 160 thread_dtor(void *mem, int size, void *arg)
 161 {
 162         struct thread *td;
 163
 164         td = (struct thread *)mem;
 165
 166 #ifdef INVARIANTS
 167         /* Verify that this thread is in a safe state to free. */
 168         switch (td->td_state) {
 169         case TDS_INHIBITED:
 170         case TDS_RUNNING:
 171         case TDS_CAN_RUN:
 172         case TDS_RUNQ:
 173                 /*
 174                  * We must never unlink a thread that is in one of
 175                  * these states, because it is currently active.
 176                  */
 177                 panic("bad state for thread unlinking");
 178                 /* NOTREACHED */
 179         case TDS_INACTIVE:
 180                 break;
 181         default:
 182                 panic("bad thread state");
 183                 /* NOTREACHED */
 184         }
 185 #endif
 186 #ifdef AUDIT
 187         audit_thread_free(td);
 188 #endif
 189         /* Free all OSD associated to this thread. */
 190         osd_thread_exit(td);
 191
 192         EVENTHANDLER_INVOKE(thread_dtor, td);
 193         tid_free(td->td_tid);
 194 }
 195
 196 /*
 197  * Initialize type-stable parts of a thread (when newly created).
 198  */
 199 static int
 200 thread_init(void *mem, int size, int flags)
 201 {
 202         struct thread *td;
 203
 204         td = (struct thread *)mem;
 205
 206         td->td_sleepqueue = sleepq_alloc();
 207         td->td_turnstile = turnstile_alloc();
 208         td->td_rlqe = NULL;
 209         EVENTHANDLER_INVOKE(thread_init, td);
 210         td->td_sched = (struct td_sched *)&td[1];
 211         umtx_thread_init(td);
 212         td->td_kstack = 0;
 213         return (0);
 214 }
 215
 216 /*
 217  * Tear down type-stable parts of a thread (just before being discarded).
 218  */
 219 static void
 220 thread_fini(void *mem, int size)
 221 {
 222         struct thread *td;
 223
 224         td = (struct thread *)mem;
 225         EVENTHANDLER_INVOKE(thread_fini, td);
 226         rlqentry_free(td->td_rlqe);
 227         turnstile_free(td->td_turnstile);
 228         sleepq_free(td->td_sleepqueue);
 229         umtx_thread_fini(td);
 230         seltdfini(td);
 231 }
 232
 233 /*
 234  * For a newly created process,
 235  * link up all the structures and its initial threads etc.
 236  * called from:
 237  * {arch}/{arch}/machdep.c   ia64_init(), init386() etc.
 238  * proc_dtor() (should go away)
 239  * proc_init()
 240  */
 241 void
 242 proc_linkup0(struct proc *p, struct thread *td)
 243 {
 244         TAILQ_INIT(&p->p_threads);           /* all threads in proc */
 245         proc_linkup(p, td);
 246 }
 247
 248 void
 249 proc_linkup(struct proc *p, struct thread *td)
 250 {
 251
 252         sigqueue_init(&p->p_sigqueue, p);
 253         p->p_ksi = ksiginfo_alloc(1);
 254         if (p->p_ksi != NULL) {
 255                 /* XXX p_ksi may be null if ksiginfo zone is not ready */
 256                 p->p_ksi->ksi_flags = KSI_EXT | KSI_INS;
 257         }
 258         LIST_INIT(&p->p_mqnotifier);
 259         p->p_numthreads = 0;
 260         thread_link(td, p);
 261 }
 262
 263 /*
 264  * Initialize global thread allocation resources.
 265  */
 266 void
 267 threadinit(void)
 268 {
 269
 270         mtx_init(&tid_lock, "TID lock", NULL, MTX_DEF);
 271
 272         /*
 273          * pid_max cannot be greater than PID_MAX.
 274          * leave one number for thread0.
 275          */
 276         tid_unrhdr = new_unrhdr(PID_MAX + 2, INT_MAX, &tid_lock);
 277
 278         thread_zone = uma_zcreate("THREAD", sched_sizeof_thread(),
 279             thread_ctor, thread_dtor, thread_init, thread_fini,
 280             16 - 1, 0);
 281         tidhashtbl = hashinit(maxproc / 2, M_TIDHASH, &tidhash);
 282         rw_init(&tidhash_lock, "tidhash");
 283 }
 284
 285 /*
 286  * Place an unused thread on the zombie list.
 287  * Use the slpq as that must be unused by now.
 288  */
 289 void
 290 thread_zombie(struct thread *td)
 291 {
 292         mtx_lock_spin(&zombie_lock);
 293         TAILQ_INSERT_HEAD(&zombie_threads, td, td_slpq);
 294         mtx_unlock_spin(&zombie_lock);
 295 }
 296
 297 /*
 298  * Release a thread that has exited after cpu_throw().
 299  */
 300 void
 301 thread_stash(struct thread *td)
 302 {
 303         atomic_subtract_rel_int(&td->td_proc->p_exitthreads, 1);
 304         thread_zombie(td);
 305 }
 306
 307 /*
 308  * Reap zombie resources.
 309  */
 310 void
 311 thread_reap(void)
 312 {
 313         struct thread *td_first, *td_next;
 314
 315         /*
 316          * Don't even bother to lock if none at this instant,
 317          * we really don't care about the next instant..
 318          */
 319         if (!TAILQ_EMPTY(&zombie_threads)) {
 320                 mtx_lock_spin(&zombie_lock);
 321                 td_first = TAILQ_FIRST(&zombie_threads);
 322                 if (td_first)
 323                         TAILQ_INIT(&zombie_threads);
 324                 mtx_unlock_spin(&zombie_lock);
 325                 while (td_first) {
 326                         td_next = TAILQ_NEXT(td_first, td_slpq);
 327                         if (td_first->td_ucred)
 328                                 crfree(td_first->td_ucred);
 329                         thread_free(td_first);
 330                         td_first = td_next;
 331                 }
 332         }
 333 }
 334
 335 /*
 336  * Allocate a thread.
 337  */
 338 struct thread *
 339 thread_alloc(int pages)
 340 {
 341         struct thread *td;
 342
 343         thread_reap(); /* check if any zombies to get */
 344
 345         td = (struct thread *)uma_zalloc(thread_zone, M_WAITOK);
 346         KASSERT(td->td_kstack == 0, ("thread_alloc got thread with kstack"));
 347         if (!vm_thread_new(td, pages)) {
 348                 uma_zfree(thread_zone, td);
 349                 return (NULL);
 350         }
 351         cpu_thread_alloc(td);
 352         return (td);
 353 }
 354
 355 int
 356 thread_alloc_stack(struct thread *td, int pages)
 357 {
 358
 359         KASSERT(td->td_kstack == 0,
 360             ("thread_alloc_stack called on a thread with kstack"));
 361         if (!vm_thread_new(td, pages))
 362                 return (0);
 363         cpu_thread_alloc(td);
 364         return (1);
 365 }
 366
 367 /*
 368  * Deallocate a thread.
 369  */
 370 void
 371 thread_free(struct thread *td)
 372 {
 373
 374         lock_profile_thread_exit(td);
 375         if (td->td_cpuset)
 376                 cpuset_rel(td->td_cpuset);
 377         td->td_cpuset = NULL;
 378         cpu_thread_free(td);
 379         if (td->td_kstack != 0)
 380                 vm_thread_dispose(td);
 381         uma_zfree(thread_zone, td);
 382 }
 383
 384 /*
 385  * Discard the current thread and exit from its context.
 386  * Always called with scheduler locked.
 387  *
 388  * Because we can't free a thread while we're operating under its context,
 389  * push the current thread into our CPU's deadthread holder. This means
 390  * we needn't worry about someone else grabbing our context before we
 391  * do a cpu_throw().
 392  */
 393 void
 394 thread_exit(void)
 395 {
 396         uint64_t runtime, new_switchtime;
 397         struct thread *td;
 398         struct thread *td2;
 399         struct proc *p;
 400         int wakeup_swapper;
 401
 402         td = curthread;
 403         p = td->td_proc;
 404
 405         PROC_SLOCK_ASSERT(p, MA_OWNED);
 406         mtx_assert(&Giant, MA_NOTOWNED);
 407
 408         PROC_LOCK_ASSERT(p, MA_OWNED);
 409         KASSERT(p != NULL, ("thread exiting without a process"));
 410         CTR3(KTR_PROC, "thread_exit: thread %p (pid %ld, %s)", td,
 411             (long)p->p_pid, td->td_name);
 412         KASSERT(TAILQ_EMPTY(&td->td_sigqueue.sq_list), ("signal pending"));
 413
 414 #ifdef AUDIT
 415         AUDIT_SYSCALL_EXIT(0, td);
 416 #endif
 417         umtx_thread_exit(td);
 418         /*
 419          * drop FPU & debug register state storage, or any other
 420          * architecture specific resources that
 421          * would not be on a new untouched process.
 422          */
 423         cpu_thread_exit(td);    /* XXXSMP */
 424
 425         /*
 426          * The last thread is left attached to the process
 427          * So that the whole bundle gets recycled. Skip
 428          * all this stuff if we never had threads.
 429          * EXIT clears all sign of other threads when
 430          * it goes to single threading, so the last thread always
 431          * takes the short path.
 432          */
 433         if (p->p_flag & P_HADTHREADS) {
 434                 if (p->p_numthreads > 1) {
 435                         thread_unlink(td);
 436                         td2 = FIRST_THREAD_IN_PROC(p);
 437                         sched_exit_thread(td2, td);
 438
 439                         /*
 440                          * The test below is NOT true if we are the
 441                          * sole exiting thread. P_STOPPED_SINGLE is unset
 442                          * in exit1() after it is the only survivor.
 443                          */
 444                         if (P_SHOULDSTOP(p) == P_STOPPED_SINGLE) {
 445                                 if (p->p_numthreads == p->p_suspcount) {
 446                                         thread_lock(p->p_singlethread);
 447                                         wakeup_swapper = thread_unsuspend_one(
 448                                                 p->p_singlethread);
 449                                         thread_unlock(p->p_singlethread);
 450                                         if (wakeup_swapper)
 451                                                 kick_proc0();
 452                                 }
 453                         }
 454
 455                         atomic_add_int(&td->td_proc->p_exitthreads, 1);
 456                         PCPU_SET(deadthread, td);
 457                 } else {
 458                         /*
 459                          * The last thread is exiting.. but not through exit()
 460                          */
 461                         panic ("thread_exit: Last thread exiting on its own");
 462                 }
 463         }
 464 #ifdef  HWPMC_HOOKS
 465         /*
 466          * If this thread is part of a process that is being tracked by hwpmc(4),
 467          * inform the module of the thread's impending exit.
 468          */
 469         if (PMC_PROC_IS_USING_PMCS(td->td_proc))
 470                 PMC_SWITCH_CONTEXT(td, PMC_FN_CSW_OUT);
 471 #endif
 472         PROC_UNLOCK(p);
 473
 474         /* Do the same timestamp bookkeeping that mi_switch() would do. */
 475         new_switchtime = cpu_ticks();
 476         runtime = new_switchtime - PCPU_GET(switchtime);
 477         td->td_runtime += runtime;
 478         td->td_incruntime += runtime;
 479         PCPU_SET(switchtime, new_switchtime);
 480         PCPU_SET(switchticks, ticks);
 481         PCPU_INC(cnt.v_swtch);
 482
 483         /* Save our resource usage in our process. */
 484         td->td_ru.ru_nvcsw++;
 485         ruxagg(p, td);
 486         rucollect(&p->p_ru, &td->td_ru);
 487
 488         thread_lock(td);
 489         PROC_SUNLOCK(p);
 490         td->td_state = TDS_INACTIVE;
 491 #ifdef WITNESS
 492         witness_thread_exit(td);
 493 #endif
 494         CTR1(KTR_PROC, "thread_exit: cpu_throw() thread %p", td);
 495         sched_throw(td);
 496         panic("I'm a teapot!");
 497         /* NOTREACHED */
 498 }
 499
 500 /*
 501  * Do any thread specific cleanups that may be needed in wait()
 502  * called with Giant, proc and schedlock not held.
 503  */
 504 void
 505 thread_wait(struct proc *p)
 506 {
 507         struct thread *td;
 508
 509         mtx_assert(&Giant, MA_NOTOWNED);
 510         KASSERT((p->p_numthreads == 1), ("Multiple threads in wait1()"));
 511         td = FIRST_THREAD_IN_PROC(p);
 512         /* Lock the last thread so we spin until it exits cpu_throw(). */
 513         thread_lock(td);
 514         thread_unlock(td);
 515         /* Wait for any remaining threads to exit cpu_throw(). */
 516         while (p->p_exitthreads)
 517                 sched_relinquish(curthread);
 518         lock_profile_thread_exit(td);
 519         cpuset_rel(td->td_cpuset);
 520         td->td_cpuset = NULL;
 521         cpu_thread_clean(td);
 522         crfree(td->td_ucred);
 523         thread_reap();  /* check for zombie threads etc. */
 524 }
 525
 526 /*
 527  * Link a thread to a process.
 528  * set up anything that needs to be initialized for it to
 529  * be used by the process.
 530  */
 531 void
 532 thread_link(struct thread *td, struct proc *p)
 533 {
 534
 535         /*
 536          * XXX This can't be enabled because it's called for proc0 before
 537          * its lock has been created.
 538          * PROC_LOCK_ASSERT(p, MA_OWNED);
 539          */
 540         td->td_state    = TDS_INACTIVE;
 541         td->td_proc     = p;
 542         td->td_flags    = TDF_INMEM;
 543
 544         LIST_INIT(&td->td_contested);
 545         LIST_INIT(&td->td_lprof[0]);
 546         LIST_INIT(&td->td_lprof[1]);
 547         sigqueue_init(&td->td_sigqueue, p);
 548         callout_init(&td->td_slpcallout, CALLOUT_MPSAFE);
 549         TAILQ_INSERT_HEAD(&p->p_threads, td, td_plist);
 550         p->p_numthreads++;
 551 }
 552
 553 /*
 554  * Convert a process with one thread to an unthreaded process.
 555  */
 556 void
 557 thread_unthread(struct thread *td)
 558 {
 559         struct proc *p = td->td_proc;
 560
 561         KASSERT((p->p_numthreads == 1), ("Unthreading with >1 threads"));
 562         p->p_flag &= ~P_HADTHREADS;
 563 }
 564
 565 /*
 566  * Called from:
 567  *  thread_exit()
 568  */
 569 void
 570 thread_unlink(struct thread *td)
 571 {
 572         struct proc *p = td->td_proc;
 573
 574         PROC_LOCK_ASSERT(p, MA_OWNED);
 575         TAILQ_REMOVE(&p->p_threads, td, td_plist);
 576         p->p_numthreads--;
 577         /* could clear a few other things here */
 578         /* Must  NOT clear links to proc! */
 579 }
 580
 581 static int
 582 calc_remaining(struct proc *p, int mode)
 583 {
 584         int remaining;
 585
 586         PROC_LOCK_ASSERT(p, MA_OWNED);
 587         PROC_SLOCK_ASSERT(p, MA_OWNED);
 588         if (mode == SINGLE_EXIT)
 589                 remaining = p->p_numthreads;
 590         else if (mode == SINGLE_BOUNDARY)
 591                 remaining = p->p_numthreads - p->p_boundary_count;
 592         else if (mode == SINGLE_NO_EXIT)
 593                 remaining = p->p_numthreads - p->p_suspcount;
 594         else
 595                 panic("calc_remaining: wrong mode %d", mode);
 596         return (remaining);
 597 }
 598
 599 /*
 600  * Enforce single-threading.
 601  *
 602  * Returns 1 if the caller must abort (another thread is waiting to
 603  * exit the process or similar). Process is locked!
 604  * Returns 0 when you are successfully the only thread running.
 605  * A process has successfully single threaded in the suspend mode when
 606  * There are no threads in user mode. Threads in the kernel must be
 607  * allowed to continue until they get to the user boundary. They may even
 608  * copy out their return values and data before suspending. They may however be
 609  * accelerated in reaching the user boundary as we will wake up
 610  * any sleeping threads that are interruptable. (PCATCH).
 611  */
 612 int
 613 thread_single(int mode)
 614 {
 615         struct thread *td;
 616         struct thread *td2;
 617         struct proc *p;
 618         int remaining, wakeup_swapper;
 619
 620         td = curthread;
 621         p = td->td_proc;
 622         mtx_assert(&Giant, MA_NOTOWNED);
 623         PROC_LOCK_ASSERT(p, MA_OWNED);
 624
 625         if ((p->p_flag & P_HADTHREADS) == 0)
 626                 return (0);
 627
 628         /* Is someone already single threading? */
 629         if (p->p_singlethread != NULL && p->p_singlethread != td)
 630                 return (1);
 631
 632         if (mode == SINGLE_EXIT) {
 633                 p->p_flag |= P_SINGLE_EXIT;
 634                 p->p_flag &= ~P_SINGLE_BOUNDARY;
 635         } else {
 636                 p->p_flag &= ~P_SINGLE_EXIT;
 637                 if (mode == SINGLE_BOUNDARY)
 638                         p->p_flag |= P_SINGLE_BOUNDARY;
 639                 else
 640                         p->p_flag &= ~P_SINGLE_BOUNDARY;
 641         }
 642         p->p_flag |= P_STOPPED_SINGLE;
 643         PROC_SLOCK(p);
 644         p->p_singlethread = td;
 645         remaining = calc_remaining(p, mode);
 646         while (remaining != 1) {
 647                 if (P_SHOULDSTOP(p) != P_STOPPED_SINGLE)
 648                         goto stopme;
 649                 wakeup_swapper = 0;
 650                 FOREACH_THREAD_IN_PROC(p, td2) {
 651                         if (td2 == td)
 652                                 continue;
 653                         thread_lock(td2);
 654                         td2->td_flags |= TDF_ASTPENDING | TDF_NEEDSUSPCHK;
 655                         if (TD_IS_INHIBITED(td2)) {
 656                                 switch (mode) {
 657                                 case SINGLE_EXIT:
 658                                         if (TD_IS_SUSPENDED(td2))
 659                                                 wakeup_swapper |=
 660                                                     thread_unsuspend_one(td2);
 661                                         if (TD_ON_SLEEPQ(td2) &&
 662                                             (td2->td_flags & TDF_SINTR))
 663                                                 wakeup_swapper |=
 664                                                     sleepq_abort(td2, EINTR);
 665                                         break;
 666                                 case SINGLE_BOUNDARY:
 667                                         if (TD_IS_SUSPENDED(td2) &&
 668                                             !(td2->td_flags & TDF_BOUNDARY))
 669                                                 wakeup_swapper |=
 670                                                     thread_unsuspend_one(td2);
 671                                         if (TD_ON_SLEEPQ(td2) &&
 672                                             (td2->td_flags & TDF_SINTR))
 673                                                 wakeup_swapper |=
 674                                                     sleepq_abort(td2, ERESTART);
 675                                         break;
 676                                 case SINGLE_NO_EXIT:
 677                                         if (TD_IS_SUSPENDED(td2) &&
 678                                             !(td2->td_flags & TDF_BOUNDARY))
 679                                                 wakeup_swapper |=
 680                                                     thread_unsuspend_one(td2);
 681                                         if (TD_ON_SLEEPQ(td2) &&
 682                                             (td2->td_flags & TDF_SINTR))
 683                                                 wakeup_swapper |=
 684                                                     sleepq_abort(td2, ERESTART);
 685                                         break;
 686                                 default:
 687                                         break;
 688                                 }
 689                         }
 690 #ifdef SMP
 691                         else if (TD_IS_RUNNING(td2) && td != td2) {
 692                                 forward_signal(td2);
 693                         }
 694 #endif
 695                         thread_unlock(td2);
 696                 }
 697                 if (wakeup_swapper)
 698                         kick_proc0();
 699                 remaining = calc_remaining(p, mode);
 700
 701                 /*
 702                  * Maybe we suspended some threads.. was it enough?
 703                  */
 704                 if (remaining == 1)
 705                         break;
 706
 707 stopme:
 708                 /*
 709                  * Wake us up when everyone else has suspended.
 710                  * In the mean time we suspend as well.
 711                  */
 712                 thread_suspend_switch(td);
 713                 remaining = calc_remaining(p, mode);
 714         }
 715         if (mode == SINGLE_EXIT) {
 716                 /*
 717                  * We have gotten rid of all the other threads and we
 718                  * are about to either exit or exec. In either case,
 719                  * we try our utmost to revert to being a non-threaded
 720                  * process.
 721                  */
 722                 p->p_singlethread = NULL;
 723                 p->p_flag &= ~(P_STOPPED_SINGLE | P_SINGLE_EXIT);
 724                 thread_unthread(td);
 725         }
 726         PROC_SUNLOCK(p);
 727         return (0);
 728 }
 729
 730 /*
 731  * Called in from locations that can safely check to see
 732  * whether we have to suspend or at least throttle for a
 733  * single-thread event (e.g. fork).
 734  *
 735  * Such locations include userret().
 736  * If the "return_instead" argument is non zero, the thread must be able to
 737  * accept 0 (caller may continue), or 1 (caller must abort) as a result.
 738  *
 739  * The 'return_instead' argument tells the function if it may do a
 740  * thread_exit() or suspend, or whether the caller must abort and back
 741  * out instead.
 742  *
 743  * If the thread that set the single_threading request has set the
 744  * P_SINGLE_EXIT bit in the process flags then this call will never return
 745  * if 'return_instead' is false, but will exit.
 746  *
 747  * P_SINGLE_EXIT | return_instead == 0| return_instead != 0
 748  *---------------+--------------------+---------------------
 749  *       0       | returns 0          |   returns 0 or 1
 750  *               | when ST ends       |   immediately
 751  *---------------+--------------------+---------------------
 752  *       1       | thread exits       |   returns 1
 753  *               |                    |  immediately
 754  * 0 = thread_exit() or suspension ok,
 755  * other = return error instead of stopping the thread.
 756  *
 757  * While a full suspension is under effect, even a single threading
 758  * thread would be suspended if it made this call (but it shouldn't).
 759  * This call should only be made from places where
 760  * thread_exit() would be safe as that may be the outcome unless
 761  * return_instead is set.
 762  */
 763 int
 764 thread_suspend_check(int return_instead)
 765 {
 766         struct thread *td;
 767         struct proc *p;
 768         int wakeup_swapper;
 769
 770         td = curthread;
 771         p = td->td_proc;
 772         mtx_assert(&Giant, MA_NOTOWNED);
 773         PROC_LOCK_ASSERT(p, MA_OWNED);
 774         while (P_SHOULDSTOP(p) ||
 775               ((p->p_flag & P_TRACED) && (td->td_dbgflags & TDB_SUSPEND))) {
 776                 if (P_SHOULDSTOP(p) == P_STOPPED_SINGLE) {
 777                         KASSERT(p->p_singlethread != NULL,
 778                             ("singlethread not set"));
 779                         /*
 780                          * The only suspension in action is a
 781                          * single-threading. Single threader need not stop.
 782                          * XXX Should be safe to access unlocked
 783                          * as it can only be set to be true by us.
 784                          */
 785                         if (p->p_singlethread == td)
 786                                 return (0);     /* Exempt from stopping. */
 787                 }
 788                 if ((p->p_flag & P_SINGLE_EXIT) && return_instead)
 789                         return (EINTR);
 790
 791                 /* Should we goto user boundary if we didn't come from there? */
 792                 if (P_SHOULDSTOP(p) == P_STOPPED_SINGLE &&
 793                     (p->p_flag & P_SINGLE_BOUNDARY) && return_instead)
 794                         return (ERESTART);
 795
 796                 /*
 797                  * Ignore suspend requests for stop signals if they
 798                  * are deferred.
 799                  */
 800                 if (P_SHOULDSTOP(p) == P_STOPPED_SIG &&
 801                     td->td_flags & TDF_SBDRY) {
 802                         KASSERT(return_instead,
 803                             ("TDF_SBDRY set for unsafe thread_suspend_check"));
 804                         return (0);
 805                 }
 806
 807                 /*
 808                  * If the process is waiting for us to exit,
 809                  * this thread should just suicide.
 810                  * Assumes that P_SINGLE_EXIT implies P_STOPPED_SINGLE.
 811                  */
 812                 if ((p->p_flag & P_SINGLE_EXIT) && (p->p_singlethread != td)) {
 813                         PROC_UNLOCK(p);
 814                         tidhash_remove(td);
 815                         PROC_LOCK(p);
 816                         tdsigcleanup(td);
 817                         PROC_SLOCK(p);
 818                         thread_stopped(p);
 819                         thread_exit();
 820                 }
 821
 822                 PROC_SLOCK(p);
 823                 thread_stopped(p);
 824                 if (P_SHOULDSTOP(p) == P_STOPPED_SINGLE) {
 825                         if (p->p_numthreads == p->p_suspcount + 1) {
 826                                 thread_lock(p->p_singlethread);
 827                                 wakeup_swapper =
 828                                     thread_unsuspend_one(p->p_singlethread);
 829                                 thread_unlock(p->p_singlethread);
 830                                 if (wakeup_swapper)
 831                                         kick_proc0();
 832                         }
 833                 }
 834                 PROC_UNLOCK(p);
 835                 thread_lock(td);
 836                 /*
 837                  * When a thread suspends, it just
 838                  * gets taken off all queues.
 839                  */
 840                 thread_suspend_one(td);
 841                 if (return_instead == 0) {
 842                         p->p_boundary_count++;
 843                         td->td_flags |= TDF_BOUNDARY;
 844                 }
 845                 PROC_SUNLOCK(p);
 846                 mi_switch(SW_INVOL | SWT_SUSPEND, NULL);
 847                 if (return_instead == 0)
 848                         td->td_flags &= ~TDF_BOUNDARY;
 849                 thread_unlock(td);
 850                 PROC_LOCK(p);
 851                 if (return_instead == 0) {
 852                         PROC_SLOCK(p);
 853                         p->p_boundary_count--;
 854                         PROC_SUNLOCK(p);
 855                 }
 856         }
 857         return (0);
 858 }
 859
 860 void
 861 thread_suspend_switch(struct thread *td)
 862 {
 863         struct proc *p;
 864
 865         p = td->td_proc;
 866         KASSERT(!TD_IS_SUSPENDED(td), ("already suspended"));
 867         PROC_LOCK_ASSERT(p, MA_OWNED);
 868         PROC_SLOCK_ASSERT(p, MA_OWNED);
 869         /*
 870          * We implement thread_suspend_one in stages here to avoid
 871          * dropping the proc lock while the thread lock is owned.
 872          */
 873         thread_stopped(p);
 874         p->p_suspcount++;
 875         PROC_UNLOCK(p);
 876         thread_lock(td);
 877         td->td_flags &= ~TDF_NEEDSUSPCHK;
 878         TD_SET_SUSPENDED(td);
 879         sched_sleep(td, 0);
 880         PROC_SUNLOCK(p);
 881         DROP_GIANT();
 882         mi_switch(SW_VOL | SWT_SUSPEND, NULL);
 883         thread_unlock(td);
 884         PICKUP_GIANT();
 885         PROC_LOCK(p);
 886         PROC_SLOCK(p);
 887 }
 888
 889 void
 890 thread_suspend_one(struct thread *td)
 891 {
 892         struct proc *p = td->td_proc;
 893
 894         PROC_SLOCK_ASSERT(p, MA_OWNED);
 895         THREAD_LOCK_ASSERT(td, MA_OWNED);
 896         KASSERT(!TD_IS_SUSPENDED(td), ("already suspended"));
 897         p->p_suspcount++;
 898         td->td_flags &= ~TDF_NEEDSUSPCHK;
 899         TD_SET_SUSPENDED(td);
 900         sched_sleep(td, 0);
 901 }
 902
 903 int
 904 thread_unsuspend_one(struct thread *td)
 905 {
 906         struct proc *p = td->td_proc;
 907
 908         PROC_SLOCK_ASSERT(p, MA_OWNED);
 909         THREAD_LOCK_ASSERT(td, MA_OWNED);
 910         KASSERT(TD_IS_SUSPENDED(td), ("Thread not suspended"));
 911         TD_CLR_SUSPENDED(td);
 912         p->p_suspcount--;
 913         return (setrunnable(td));
 914 }
 915
 916 /*
 917  * Allow all threads blocked by single threading to continue running.
 918  */
 919 void
 920 thread_unsuspend(struct proc *p)
 921 {
 922         struct thread *td;
 923         int wakeup_swapper;
 924
 925         PROC_LOCK_ASSERT(p, MA_OWNED);
 926         PROC_SLOCK_ASSERT(p, MA_OWNED);
 927         wakeup_swapper = 0;
 928         if (!P_SHOULDSTOP(p)) {
 929                 FOREACH_THREAD_IN_PROC(p, td) {
 930                         thread_lock(td);
 931                         if (TD_IS_SUSPENDED(td)) {
 932                                 wakeup_swapper |= thread_unsuspend_one(td);
 933                         }
 934                         thread_unlock(td);
 935                 }
 936         } else if ((P_SHOULDSTOP(p) == P_STOPPED_SINGLE) &&
 937             (p->p_numthreads == p->p_suspcount)) {
 938                 /*
 939                  * Stopping everything also did the job for the single
 940                  * threading request. Now we've downgraded to single-threaded,
 941                  * let it continue.
 942                  */
 943                 thread_lock(p->p_singlethread);
 944                 wakeup_swapper = thread_unsuspend_one(p->p_singlethread);
 945                 thread_unlock(p->p_singlethread);
 946         }
 947         if (wakeup_swapper)
 948                 kick_proc0();
 949 }
 950
 951 /*
 952  * End the single threading mode..
 953  */
 954 void
 955 thread_single_end(void)
 956 {
 957         struct thread *td;
 958         struct proc *p;
 959         int wakeup_swapper;
 960
 961         td = curthread;
 962         p = td->td_proc;
 963         PROC_LOCK_ASSERT(p, MA_OWNED);
 964         p->p_flag &= ~(P_STOPPED_SINGLE | P_SINGLE_EXIT | P_SINGLE_BOUNDARY);
 965         PROC_SLOCK(p);
 966         p->p_singlethread = NULL;
 967         wakeup_swapper = 0;
 968         /*
 969          * If there are other threads they may now run,
 970          * unless of course there is a blanket 'stop order'
 971          * on the process. The single threader must be allowed
 972          * to continue however as this is a bad place to stop.
 973          */
 974         if ((p->p_numthreads != 1) && (!P_SHOULDSTOP(p))) {
 975                 FOREACH_THREAD_IN_PROC(p, td) {
 976                         thread_lock(td);
 977                         if (TD_IS_SUSPENDED(td)) {
 978                                 wakeup_swapper |= thread_unsuspend_one(td);
 979                         }
 980                         thread_unlock(td);
 981                 }
 982         }
 983         PROC_SUNLOCK(p);
 984         if (wakeup_swapper)
 985                 kick_proc0();
 986 }
 987
 988 struct thread *
 989 thread_find(struct proc *p, lwpid_t tid)
 990 {
 991         struct thread *td;
 992
 993         PROC_LOCK_ASSERT(p, MA_OWNED);
 994         FOREACH_THREAD_IN_PROC(p, td) {
 995                 if (td->td_tid == tid)
 996                         break;
 997         }
 998         return (td);
 999 }
1000
1001 /* Locate a thread by number; return with proc lock held. */
1002 struct thread *
1003 tdfind(lwpid_t tid, pid_t pid)
1004 {
1005 #define RUN_THRESH      16
1006         struct thread *td;
1007         int run = 0;
1008
1009         rw_rlock(&tidhash_lock);
1010         LIST_FOREACH(td, TIDHASH(tid), td_hash) {
1011                 if (td->td_tid == tid) {
1012                         if (pid != -1 && td->td_proc->p_pid != pid) {
1013                                 td = NULL;
1014                                 break;
1015                         }
1016                         PROC_LOCK(td->td_proc);
1017                         if (td->td_proc->p_state == PRS_NEW) {
1018                                 PROC_UNLOCK(td->td_proc);
1019                                 td = NULL;
1020                                 break;
1021                         }
1022                         if (run > RUN_THRESH) {
1023                                 if (rw_try_upgrade(&tidhash_lock)) {
1024                                         LIST_REMOVE(td, td_hash);
1025                                         LIST_INSERT_HEAD(TIDHASH(td->td_tid),
1026                                                 td, td_hash);
1027                                         rw_wunlock(&tidhash_lock);
1028                                         return (td);
1029                                 }
1030                         }
1031                         break;
1032                 }
1033                 run++;
1034         }
1035         rw_runlock(&tidhash_lock);
1036         return (td);
1037 }
1038
1039 void
1040 tidhash_add(struct thread *td)
1041 {
1042         rw_wlock(&tidhash_lock);
1043         LIST_INSERT_HEAD(TIDHASH(td->td_tid), td, td_hash);
1044         rw_wunlock(&tidhash_lock);
1045 }
1046
1047 void
1048 tidhash_remove(struct thread *td)
1049 {
1050         rw_wlock(&tidhash_lock);
1051         LIST_REMOVE(td, td_hash);
1052         rw_wunlock(&tidhash_lock);
1053 }