kern_exit.c revision 1.120 1 1.120 yamt /* $NetBSD: kern_exit.c,v 1.120 2003/08/20 13:50:53 yamt Exp $ */
2 1.56 thorpej
3 1.56 thorpej /*-
4 1.72 thorpej * Copyright (c) 1998, 1999 The NetBSD Foundation, Inc.
5 1.56 thorpej * All rights reserved.
6 1.58 christos *
7 1.56 thorpej * This code is derived from software contributed to The NetBSD Foundation
8 1.56 thorpej * by Jason R. Thorpe of the Numerical Aerospace Simulation Facility,
9 1.58 christos * NASA Ames Research Center.
10 1.56 thorpej *
11 1.56 thorpej * Redistribution and use in source and binary forms, with or without
12 1.56 thorpej * modification, are permitted provided that the following conditions
13 1.56 thorpej * are met:
14 1.56 thorpej * 1. Redistributions of source code must retain the above copyright
15 1.56 thorpej * notice, this list of conditions and the following disclaimer.
16 1.58 christos * 2. Redistributions in binary form must reproduce the above copyright
17 1.56 thorpej * notice, this list of conditions and the following disclaimer in the
18 1.56 thorpej * documentation and/or other materials provided with the distribution.
19 1.56 thorpej * 3. All advertising materials mentioning features or use of this software
20 1.56 thorpej * must display the following acknowledgement:
21 1.58 christos * This product includes software developed by the NetBSD
22 1.58 christos * Foundation, Inc. and its contributors.
23 1.56 thorpej * 4. Neither the name of The NetBSD Foundation nor the names of its
24 1.56 thorpej * contributors may be used to endorse or promote products derived
25 1.56 thorpej * from this software without specific prior written permission.
26 1.58 christos *
27 1.56 thorpej * THIS SOFTWARE IS PROVIDED BY THE NETBSD FOUNDATION, INC. AND CONTRIBUTORS
28 1.56 thorpej * ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
29 1.56 thorpej * TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
30 1.56 thorpej * PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE FOUNDATION OR CONTRIBUTORS
31 1.56 thorpej * BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
32 1.56 thorpej * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
33 1.56 thorpej * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
34 1.56 thorpej * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
35 1.56 thorpej * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
36 1.56 thorpej * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
37 1.56 thorpej * POSSIBILITY OF SUCH DAMAGE.
38 1.56 thorpej */
39 1.24 cgd
40 1.24 cgd /*
41 1.24 cgd * Copyright (c) 1982, 1986, 1989, 1991, 1993
42 1.24 cgd * The Regents of the University of California. All rights reserved.
43 1.24 cgd * (c) UNIX System Laboratories, Inc.
44 1.24 cgd * All or some portions of this file are derived from material licensed
45 1.24 cgd * to the University of California by American Telephone and Telegraph
46 1.24 cgd * Co. or Unix System Laboratories, Inc. and are reproduced herein with
47 1.24 cgd * the permission of UNIX System Laboratories, Inc.
48 1.24 cgd *
49 1.24 cgd * Redistribution and use in source and binary forms, with or without
50 1.24 cgd * modification, are permitted provided that the following conditions
51 1.24 cgd * are met:
52 1.24 cgd * 1. Redistributions of source code must retain the above copyright
53 1.24 cgd * notice, this list of conditions and the following disclaimer.
54 1.24 cgd * 2. Redistributions in binary form must reproduce the above copyright
55 1.24 cgd * notice, this list of conditions and the following disclaimer in the
56 1.24 cgd * documentation and/or other materials provided with the distribution.
57 1.119 agc * 3. Neither the name of the University nor the names of its contributors
58 1.24 cgd * may be used to endorse or promote products derived from this software
59 1.24 cgd * without specific prior written permission.
60 1.24 cgd *
61 1.24 cgd * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
62 1.24 cgd * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
63 1.24 cgd * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
64 1.24 cgd * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
65 1.24 cgd * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
66 1.24 cgd * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
67 1.24 cgd * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
68 1.24 cgd * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
69 1.24 cgd * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
70 1.24 cgd * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
71 1.24 cgd * SUCH DAMAGE.
72 1.24 cgd *
73 1.49 fvdl * @(#)kern_exit.c 8.10 (Berkeley) 2/23/95
74 1.24 cgd */
75 1.92 lukem
76 1.92 lukem #include <sys/cdefs.h>
77 1.120 yamt __KERNEL_RCSID(0, "$NetBSD: kern_exit.c,v 1.120 2003/08/20 13:50:53 yamt Exp $");
78 1.48 mrg
79 1.51 thorpej #include "opt_ktrace.h"
80 1.97 briggs #include "opt_perfctrs.h"
81 1.94 christos #include "opt_systrace.h"
82 1.60 tron #include "opt_sysv.h"
83 1.24 cgd
84 1.24 cgd #include <sys/param.h>
85 1.24 cgd #include <sys/systm.h>
86 1.24 cgd #include <sys/ioctl.h>
87 1.24 cgd #include <sys/proc.h>
88 1.24 cgd #include <sys/tty.h>
89 1.24 cgd #include <sys/time.h>
90 1.24 cgd #include <sys/resource.h>
91 1.24 cgd #include <sys/kernel.h>
92 1.50 christos #include <sys/ktrace.h>
93 1.24 cgd #include <sys/proc.h>
94 1.24 cgd #include <sys/buf.h>
95 1.24 cgd #include <sys/wait.h>
96 1.24 cgd #include <sys/file.h>
97 1.24 cgd #include <sys/vnode.h>
98 1.24 cgd #include <sys/syslog.h>
99 1.24 cgd #include <sys/malloc.h>
100 1.53 thorpej #include <sys/pool.h>
101 1.24 cgd #include <sys/resourcevar.h>
102 1.98 briggs #if defined(PERFCTRS)
103 1.97 briggs #include <sys/pmc.h>
104 1.98 briggs #endif
105 1.24 cgd #include <sys/ptrace.h>
106 1.29 cgd #include <sys/acct.h>
107 1.36 christos #include <sys/filedesc.h>
108 1.100 gmcgarry #include <sys/ras.h>
109 1.36 christos #include <sys/signalvar.h>
110 1.64 ross #include <sys/sched.h>
111 1.107 thorpej #include <sys/sa.h>
112 1.107 thorpej #include <sys/savar.h>
113 1.26 cgd #include <sys/mount.h>
114 1.26 cgd #include <sys/syscallargs.h>
115 1.94 christos #include <sys/systrace.h>
116 1.26 cgd
117 1.24 cgd #include <machine/cpu.h>
118 1.24 cgd
119 1.47 mrg #include <uvm/uvm_extern.h>
120 1.47 mrg
121 1.107 thorpej #define DEBUG_EXIT
122 1.107 thorpej
123 1.107 thorpej #ifdef DEBUG_EXIT
124 1.107 thorpej int debug_exit = 0;
125 1.107 thorpej #define DPRINTF(x) if (debug_exit) printf x
126 1.107 thorpej #else
127 1.107 thorpej #define DPRINTF(x)
128 1.107 thorpej #endif
129 1.107 thorpej
130 1.107 thorpej static void lwp_exit_hook(struct lwp *, void *);
131 1.94 christos
132 1.24 cgd /*
133 1.24 cgd * exit --
134 1.24 cgd * Death of process.
135 1.24 cgd */
136 1.31 thorpej int
137 1.107 thorpej sys_exit(struct lwp *l, void *v, register_t *retval)
138 1.31 thorpej {
139 1.33 mycroft struct sys_exit_args /* {
140 1.89 lukem syscallarg(int) rval;
141 1.31 thorpej } */ *uap = v;
142 1.24 cgd
143 1.107 thorpej /* Don't call exit1() multiple times in the same process.*/
144 1.107 thorpej if (l->l_proc->p_flag & P_WEXIT)
145 1.107 thorpej lwp_exit(l);
146 1.107 thorpej
147 1.107 thorpej exit1(l, W_EXITCODE(SCARG(uap, rval), 0));
148 1.24 cgd /* NOTREACHED */
149 1.31 thorpej return (0);
150 1.24 cgd }
151 1.24 cgd
152 1.24 cgd /*
153 1.24 cgd * Exit: deallocate address space and other resources, change proc state
154 1.24 cgd * to zombie, and unlink proc from allproc and parent's lists. Save exit
155 1.24 cgd * status and rusage for wait(). Check for child processes and orphan them.
156 1.24 cgd */
157 1.31 thorpej void
158 1.107 thorpej exit1(struct lwp *l, int rv)
159 1.24 cgd {
160 1.107 thorpej struct proc *p, *q, *nq;
161 1.108 nathanw int s, sa;
162 1.24 cgd
163 1.107 thorpej p = l->l_proc;
164 1.107 thorpej
165 1.78 thorpej if (__predict_false(p == initproc))
166 1.24 cgd panic("init died (signal %d, exit %d)",
167 1.24 cgd WTERMSIG(rv), WEXITSTATUS(rv));
168 1.73 thorpej
169 1.112 nathanw p->p_flag |= P_WEXIT;
170 1.112 nathanw
171 1.107 thorpej DPRINTF(("exit1: %d.%d exiting.\n", p->p_pid, l->l_lid));
172 1.107 thorpej /*
173 1.107 thorpej * Disable scheduler activation upcalls.
174 1.107 thorpej * We're trying to get out of here.
175 1.107 thorpej */
176 1.108 nathanw sa = 0;
177 1.110 nathanw if (p->p_sa != NULL) {
178 1.118 fvdl
179 1.107 thorpej l->l_flag &= ~L_SA;
180 1.118 fvdl #if 0
181 1.107 thorpej p->p_flag &= ~P_SA;
182 1.118 fvdl #endif
183 1.108 nathanw sa = 1;
184 1.107 thorpej }
185 1.107 thorpej
186 1.24 cgd #ifdef PGINPROF
187 1.24 cgd vmsizmon();
188 1.24 cgd #endif
189 1.24 cgd if (p->p_flag & P_PROFIL)
190 1.24 cgd stopprofclock(p);
191 1.55 thorpej p->p_ru = pool_get(&rusage_pool, PR_WAITOK);
192 1.24 cgd /*
193 1.37 mycroft * If parent is waiting for us to exit or exec, P_PPWAIT is set; we
194 1.37 mycroft * wake up the parent early to avoid deadlock.
195 1.24 cgd */
196 1.37 mycroft if (p->p_flag & P_PPWAIT) {
197 1.37 mycroft p->p_flag &= ~P_PPWAIT;
198 1.37 mycroft wakeup((caddr_t)p->p_pptr);
199 1.37 mycroft }
200 1.87 jdolecek sigfillset(&p->p_sigctx.ps_sigignore);
201 1.87 jdolecek sigemptyset(&p->p_sigctx.ps_siglist);
202 1.87 jdolecek p->p_sigctx.ps_sigcheck = 0;
203 1.107 thorpej timers_free(p, TIMERS_ALL);
204 1.107 thorpej
205 1.108 nathanw if (sa || (p->p_nlwps > 1))
206 1.107 thorpej exit_lwps(l);
207 1.100 gmcgarry
208 1.100 gmcgarry #if defined(__HAVE_RAS)
209 1.100 gmcgarry ras_purgeall(p);
210 1.100 gmcgarry #endif
211 1.24 cgd
212 1.24 cgd /*
213 1.24 cgd * Close open files and release open-file table.
214 1.24 cgd * This may block!
215 1.24 cgd */
216 1.117 fvdl fdfree(p);
217 1.66 thorpej cwdfree(p);
218 1.24 cgd
219 1.93 christos doexithooks(p);
220 1.93 christos
221 1.24 cgd if (SESS_LEADER(p)) {
222 1.76 augustss struct session *sp = p->p_session;
223 1.24 cgd
224 1.24 cgd if (sp->s_ttyvp) {
225 1.24 cgd /*
226 1.24 cgd * Controlling process.
227 1.24 cgd * Signal foreground pgrp,
228 1.24 cgd * drain controlling terminal
229 1.24 cgd * and revoke access to controlling terminal.
230 1.24 cgd */
231 1.24 cgd if (sp->s_ttyp->t_session == sp) {
232 1.24 cgd if (sp->s_ttyp->t_pgrp)
233 1.24 cgd pgsignal(sp->s_ttyp->t_pgrp, SIGHUP, 1);
234 1.24 cgd (void) ttywait(sp->s_ttyp);
235 1.24 cgd /*
236 1.24 cgd * The tty could have been revoked
237 1.24 cgd * if we blocked.
238 1.24 cgd */
239 1.24 cgd if (sp->s_ttyvp)
240 1.49 fvdl VOP_REVOKE(sp->s_ttyvp, REVOKEALL);
241 1.24 cgd }
242 1.24 cgd if (sp->s_ttyvp)
243 1.24 cgd vrele(sp->s_ttyvp);
244 1.24 cgd sp->s_ttyvp = NULL;
245 1.24 cgd /*
246 1.24 cgd * s_ttyp is not zero'd; we use this to indicate
247 1.24 cgd * that the session once had a controlling terminal.
248 1.24 cgd * (for logging and informational purposes)
249 1.24 cgd */
250 1.24 cgd }
251 1.24 cgd sp->s_leader = NULL;
252 1.24 cgd }
253 1.24 cgd fixjobc(p, p->p_pgrp, 0);
254 1.117 fvdl (void)acct_process(p);
255 1.24 cgd #ifdef KTRACE
256 1.24 cgd /*
257 1.24 cgd * release trace file
258 1.24 cgd */
259 1.50 christos ktrderef(p);
260 1.94 christos #endif
261 1.94 christos #ifdef SYSTRACE
262 1.94 christos systrace_sys_exit(p);
263 1.24 cgd #endif
264 1.24 cgd /*
265 1.99 manu * If emulation has process exit hook, call it now.
266 1.99 manu */
267 1.99 manu if (p->p_emul->e_proc_exit)
268 1.99 manu (*p->p_emul->e_proc_exit)(p);
269 1.99 manu
270 1.99 manu /*
271 1.72 thorpej * Give orphaned children to init(8).
272 1.72 thorpej */
273 1.101 matt q = LIST_FIRST(&p->p_children);
274 1.25 mycroft if (q) /* only need this if any child is S_ZOMB */
275 1.37 mycroft wakeup((caddr_t)initproc);
276 1.25 mycroft for (; q != 0; q = nq) {
277 1.101 matt nq = LIST_NEXT(q, p_sibling);
278 1.104 jdolecek
279 1.24 cgd /*
280 1.104 jdolecek * Traced processes are killed since their existence
281 1.104 jdolecek * means someone is screwing up. Since we reset the
282 1.104 jdolecek * trace flags, the logic in sys_wait4() would not be
283 1.104 jdolecek * triggered to reparent the process to its
284 1.106 jdolecek * original parent, so we must do this here.
285 1.24 cgd */
286 1.24 cgd if (q->p_flag & P_TRACED) {
287 1.104 jdolecek if (q->p_opptr != q->p_pptr) {
288 1.104 jdolecek struct proc *t = q->p_opptr;
289 1.104 jdolecek proc_reparent(q, t ? t : initproc);
290 1.104 jdolecek q->p_opptr = NULL;
291 1.105 jdolecek } else
292 1.105 jdolecek proc_reparent(q, initproc);
293 1.45 mycroft q->p_flag &= ~(P_TRACED|P_WAITED|P_FSTRACE);
294 1.24 cgd psignal(q, SIGKILL);
295 1.104 jdolecek } else {
296 1.104 jdolecek proc_reparent(q, initproc);
297 1.24 cgd }
298 1.24 cgd }
299 1.24 cgd
300 1.24 cgd /*
301 1.95 jdolecek * Reset p_opptr pointer of all former children which got
302 1.95 jdolecek * traced by another process and were reparented. We reset
303 1.95 jdolecek * it to NULL here; the trace detach code then reparents
304 1.95 jdolecek * the child to initproc. We only check allproc list, since
305 1.95 jdolecek * eventual former children on zombproc list won't reference
306 1.95 jdolecek * p_opptr anymore.
307 1.95 jdolecek */
308 1.95 jdolecek if (p->p_flag & P_CHTRACED) {
309 1.95 jdolecek struct proc *t;
310 1.95 jdolecek
311 1.95 jdolecek proclist_lock_read();
312 1.95 jdolecek
313 1.95 jdolecek LIST_FOREACH(t, &allproc, p_list) {
314 1.95 jdolecek if (t->p_opptr == p)
315 1.95 jdolecek t->p_opptr = NULL;
316 1.95 jdolecek }
317 1.95 jdolecek
318 1.95 jdolecek proclist_unlock_read();
319 1.95 jdolecek }
320 1.103 jdolecek
321 1.103 jdolecek /*
322 1.115 dsl * Save exit status and final rusage info, adding in child rusage
323 1.115 dsl * info and self times.
324 1.115 dsl * In order to pick up the time for the current execution, we must
325 1.115 dsl * do this before unlinking the lwp from l_list.
326 1.115 dsl */
327 1.115 dsl p->p_xstat = rv;
328 1.115 dsl *p->p_ru = p->p_stats->p_ru;
329 1.115 dsl calcru(p, &p->p_ru->ru_utime, &p->p_ru->ru_stime, NULL);
330 1.115 dsl ruadd(p->p_ru, &p->p_stats->p_cru);
331 1.115 dsl
332 1.115 dsl /*
333 1.115 dsl * NOTE: WE ARE NO LONGER ALLOWED TO SLEEP!
334 1.115 dsl */
335 1.115 dsl
336 1.115 dsl /*
337 1.115 dsl * Move proc from allproc to zombproc, but do not yet
338 1.115 dsl * wake up the reaper. We will put the proc on the
339 1.115 dsl * deadproc list later (using the p_dead member), and
340 1.115 dsl * wake up the reaper when we do.
341 1.115 dsl * Changing the state to SDEAD stops it being found by pfind().
342 1.115 dsl */
343 1.115 dsl s = proclist_lock_write();
344 1.115 dsl p->p_stat = SDEAD;
345 1.115 dsl p->p_nrlwps--;
346 1.115 dsl l->l_stat = SDEAD;
347 1.115 dsl LIST_REMOVE(p, p_list);
348 1.115 dsl LIST_INSERT_HEAD(&zombproc, p, p_list);
349 1.115 dsl LIST_REMOVE(l, l_list);
350 1.115 dsl l->l_flag |= L_DETACHED;
351 1.115 dsl proclist_unlock_write(s);
352 1.115 dsl
353 1.115 dsl /*
354 1.103 jdolecek * Notify interested parties of our demise.
355 1.103 jdolecek */
356 1.103 jdolecek KNOTE(&p->p_klist, NOTE_EXIT);
357 1.97 briggs
358 1.97 briggs #if PERFCTRS
359 1.97 briggs /*
360 1.97 briggs * Save final PMC information in parent process & clean up.
361 1.97 briggs */
362 1.97 briggs if (PMC_ENABLED(p)) {
363 1.97 briggs pmc_save_context(p);
364 1.97 briggs pmc_accumulate(p->p_pptr, p);
365 1.97 briggs pmc_process_exit(p);
366 1.97 briggs }
367 1.97 briggs #endif
368 1.59 christos
369 1.59 christos /*
370 1.59 christos * Notify parent that we're gone. If parent has the P_NOCLDWAIT
371 1.59 christos * flag set, notify init instead (and hope it will handle
372 1.59 christos * this situation).
373 1.59 christos */
374 1.59 christos if (p->p_pptr->p_flag & P_NOCLDWAIT) {
375 1.59 christos struct proc *pp = p->p_pptr;
376 1.59 christos proc_reparent(p, initproc);
377 1.59 christos /*
378 1.59 christos * If this was the last child of our parent, notify
379 1.59 christos * parent, so in case he was wait(2)ing, he will
380 1.59 christos * continue.
381 1.59 christos */
382 1.101 matt if (LIST_FIRST(&pp->p_children) == NULL)
383 1.59 christos wakeup((caddr_t)pp);
384 1.59 christos }
385 1.67 thorpej
386 1.67 thorpej /*
387 1.67 thorpej * Release the process's signal state.
388 1.67 thorpej */
389 1.67 thorpej sigactsfree(p);
390 1.96 manu
391 1.96 manu /*
392 1.107 thorpej * Clear curlwp after we've done all operations
393 1.24 cgd * that could block, and before tearing down the rest
394 1.24 cgd * of the process state that might be used from clock, etc.
395 1.107 thorpej * Also, can't clear curlwp while we're still runnable,
396 1.24 cgd * as we're not on a run queue (we are current, just not
397 1.24 cgd * a proper proc any longer!).
398 1.24 cgd *
399 1.24 cgd * Other substructures are freed from wait().
400 1.24 cgd */
401 1.107 thorpej curlwp = NULL;
402 1.74 bouyer limfree(p->p_limit);
403 1.107 thorpej pstatsfree(p->p_stats);
404 1.77 thorpej p->p_limit = NULL;
405 1.24 cgd
406 1.85 thorpej /* This process no longer needs to hold the kernel lock. */
407 1.107 thorpej KERNEL_PROC_UNLOCK(l);
408 1.85 thorpej
409 1.24 cgd /*
410 1.56 thorpej * Finally, call machine-dependent code to switch to a new
411 1.56 thorpej * context (possibly the idle context). Once we are no longer
412 1.56 thorpej * using the dead process's vmspace and stack, exit2() will be
413 1.56 thorpej * called to schedule those resources to be released by the
414 1.56 thorpej * reaper thread.
415 1.56 thorpej *
416 1.56 thorpej * Note that cpu_exit() will end with a call equivalent to
417 1.56 thorpej * cpu_switch(), finishing our execution (pun intended).
418 1.24 cgd */
419 1.107 thorpej
420 1.107 thorpej cpu_exit(l, 1);
421 1.107 thorpej }
422 1.107 thorpej
423 1.107 thorpej void
424 1.107 thorpej exit_lwps(struct lwp *l)
425 1.107 thorpej {
426 1.107 thorpej struct proc *p;
427 1.107 thorpej struct lwp *l2;
428 1.107 thorpej int s, error;
429 1.107 thorpej lwpid_t waited;
430 1.107 thorpej
431 1.107 thorpej p = l->l_proc;
432 1.107 thorpej
433 1.107 thorpej /* XXX SMP
434 1.107 thorpej * This would be the right place to IPI any LWPs running on
435 1.107 thorpej * other processors so that they can notice the userret exit hook.
436 1.107 thorpej */
437 1.107 thorpej p->p_userret = lwp_exit_hook;
438 1.107 thorpej p->p_userret_arg = NULL;
439 1.107 thorpej
440 1.107 thorpej /*
441 1.107 thorpej * Make SA-cached LWPs normal process runnable LWPs so that
442 1.107 thorpej * they'll also self-destruct.
443 1.107 thorpej */
444 1.107 thorpej if (p->p_sa && p->p_sa->sa_ncached > 0) {
445 1.107 thorpej DPRINTF(("exit_lwps: Making cached LWPs of %d runnable: ",
446 1.107 thorpej p->p_pid));
447 1.107 thorpej SCHED_LOCK(s);
448 1.107 thorpej while ((l2 = sa_getcachelwp(p)) != 0) {
449 1.107 thorpej l2->l_priority = l2->l_usrpri;
450 1.107 thorpej setrunnable(l2);
451 1.107 thorpej DPRINTF(("%d ", l2->l_lid));
452 1.107 thorpej }
453 1.107 thorpej DPRINTF(("\n"));
454 1.107 thorpej SCHED_UNLOCK(s);
455 1.107 thorpej }
456 1.107 thorpej
457 1.107 thorpej /*
458 1.107 thorpej * Interrupt LWPs in interruptable sleep, unsuspend suspended
459 1.111 skrll * LWPs, make detached LWPs undetached (so we can wait for
460 1.107 thorpej * them) and then wait for everyone else to finish.
461 1.107 thorpej */
462 1.107 thorpej LIST_FOREACH(l2, &p->p_lwps, l_sibling) {
463 1.118 fvdl #if 0
464 1.107 thorpej l2->l_flag &= ~(L_DETACHED|L_SA);
465 1.118 fvdl #endif
466 1.118 fvdl l2->l_flag &= ~(L_DETACHED);
467 1.118 fvdl
468 1.118 fvdl if(l2->l_flag & L_SA_WANTS_VP)
469 1.118 fvdl {
470 1.118 fvdl wakeup(l2);
471 1.118 fvdl }
472 1.118 fvdl
473 1.107 thorpej if ((l2->l_stat == LSSLEEP && (l2->l_flag & L_SINTR)) ||
474 1.118 fvdl l2->l_stat == LSSUSPENDED || l2->l_stat == LSSTOP) {
475 1.107 thorpej SCHED_LOCK(s);
476 1.107 thorpej setrunnable(l2);
477 1.107 thorpej SCHED_UNLOCK(s);
478 1.107 thorpej DPRINTF(("exit_lwps: Made %d.%d runnable\n",
479 1.107 thorpej p->p_pid, l2->l_lid));
480 1.107 thorpej }
481 1.107 thorpej }
482 1.107 thorpej
483 1.107 thorpej
484 1.107 thorpej while (p->p_nlwps > 1) {
485 1.107 thorpej DPRINTF(("exit_lwps: waiting for %d LWPs (%d runnable, %d zombies)\n",
486 1.107 thorpej p->p_nlwps, p->p_nrlwps, p->p_nzlwps));
487 1.107 thorpej error = lwp_wait1(l, 0, &waited, LWPWAIT_EXITCONTROL);
488 1.107 thorpej if (error)
489 1.107 thorpej panic("exit_lwps: lwp_wait1 failed with error %d\n",
490 1.107 thorpej error);
491 1.107 thorpej DPRINTF(("exit_lwps: Got LWP %d from lwp_wait1()\n", waited));
492 1.107 thorpej }
493 1.107 thorpej
494 1.107 thorpej p->p_userret = NULL;
495 1.107 thorpej }
496 1.107 thorpej
497 1.107 thorpej /* Wrapper function for use in p_userret */
498 1.107 thorpej static void
499 1.107 thorpej lwp_exit_hook(struct lwp *l, void *arg)
500 1.107 thorpej {
501 1.107 thorpej KERNEL_PROC_LOCK(l);
502 1.107 thorpej lwp_exit(l);
503 1.24 cgd }
504 1.24 cgd
505 1.56 thorpej /*
506 1.56 thorpej * We are called from cpu_exit() once it is safe to schedule the
507 1.85 thorpej * dead process's resources to be freed (i.e., once we've switched to
508 1.85 thorpej * the idle PCB for the current CPU).
509 1.72 thorpej *
510 1.72 thorpej * NOTE: One must be careful with locking in this routine. It's
511 1.72 thorpej * called from a critical section in machine-dependent code, so
512 1.72 thorpej * we should refrain from changing any interrupt state.
513 1.72 thorpej *
514 1.72 thorpej * We lock the deadproc list (a spin lock), place the proc on that
515 1.115 dsl * list (using the p_dead member), and wake up the reaper.
516 1.56 thorpej */
517 1.56 thorpej void
518 1.107 thorpej exit2(struct lwp *l)
519 1.56 thorpej {
520 1.107 thorpej struct proc *p = l->l_proc;
521 1.56 thorpej
522 1.72 thorpej simple_lock(&deadproc_slock);
523 1.115 dsl SLIST_INSERT_HEAD(&deadprocs, p, p_dead);
524 1.72 thorpej simple_unlock(&deadproc_slock);
525 1.72 thorpej
526 1.107 thorpej /* lwp_exit2() will wake up deadproc for us. */
527 1.107 thorpej lwp_exit2(l);
528 1.56 thorpej }
529 1.56 thorpej
530 1.56 thorpej /*
531 1.56 thorpej * Process reaper. This is run by a kernel thread to free the resources
532 1.56 thorpej * of a dead process. Once the resources are free, the process becomes
533 1.56 thorpej * a zombie, and the parent is allowed to read the undead's status.
534 1.56 thorpej */
535 1.56 thorpej void
536 1.84 thorpej reaper(void *arg)
537 1.56 thorpej {
538 1.56 thorpej struct proc *p;
539 1.107 thorpej struct lwp *l;
540 1.56 thorpej
541 1.107 thorpej KERNEL_PROC_UNLOCK(curlwp);
542 1.90 fvdl
543 1.56 thorpej for (;;) {
544 1.72 thorpej simple_lock(&deadproc_slock);
545 1.115 dsl p = SLIST_FIRST(&deadprocs);
546 1.107 thorpej l = LIST_FIRST(&deadlwp);
547 1.107 thorpej if (p == NULL && l == NULL) {
548 1.56 thorpej /* No work for us; go to sleep until someone exits. */
549 1.115 dsl (void) ltsleep(&deadprocs, PVM|PNORELOCK,
550 1.83 thorpej "reaper", 0, &deadproc_slock);
551 1.56 thorpej continue;
552 1.56 thorpej }
553 1.56 thorpej
554 1.107 thorpej if (l != NULL ) {
555 1.107 thorpej p = l->l_proc;
556 1.56 thorpej
557 1.115 dsl /* Remove lwp from the deadlwp list. */
558 1.107 thorpej LIST_REMOVE(l, l_list);
559 1.107 thorpej simple_unlock(&deadproc_slock);
560 1.107 thorpej KERNEL_PROC_LOCK(curlwp);
561 1.107 thorpej
562 1.107 thorpej /*
563 1.107 thorpej * Give machine-dependent code a chance to free any
564 1.107 thorpej * resources it couldn't free while still running on
565 1.107 thorpej * that process's context. This must be done before
566 1.107 thorpej * uvm_lwp_exit(), in case these resources are in the
567 1.107 thorpej * PCB.
568 1.107 thorpej */
569 1.107 thorpej cpu_wait(l);
570 1.71 thorpej
571 1.107 thorpej /*
572 1.107 thorpej * Free the VM resources we're still holding on to.
573 1.107 thorpej */
574 1.107 thorpej uvm_lwp_exit(l);
575 1.56 thorpej
576 1.107 thorpej l->l_stat = LSZOMB;
577 1.107 thorpej if (l->l_flag & L_DETACHED) {
578 1.107 thorpej /* Nobody waits for detached LWPs. */
579 1.107 thorpej LIST_REMOVE(l, l_sibling);
580 1.107 thorpej p->p_nlwps--;
581 1.107 thorpej pool_put(&lwp_pool, l);
582 1.107 thorpej } else {
583 1.107 thorpej p->p_nzlwps++;
584 1.107 thorpej wakeup((caddr_t)&p->p_nlwps);
585 1.107 thorpej }
586 1.107 thorpej /* XXXNJW where should this be with respect to
587 1.107 thorpej * the wakeup() above? */
588 1.107 thorpej KERNEL_PROC_UNLOCK(curlwp);
589 1.107 thorpej } else {
590 1.115 dsl /* Remove proc from the deadproc list. */
591 1.115 dsl SLIST_REMOVE_HEAD(&deadprocs, p_dead);
592 1.107 thorpej simple_unlock(&deadproc_slock);
593 1.107 thorpej KERNEL_PROC_LOCK(curlwp);
594 1.56 thorpej
595 1.107 thorpej /*
596 1.107 thorpej * Free the VM resources we're still holding on to.
597 1.107 thorpej * We must do this from a valid thread because doing
598 1.107 thorpej * so may block.
599 1.107 thorpej */
600 1.107 thorpej uvm_proc_exit(p);
601 1.107 thorpej
602 1.107 thorpej /* Process is now a true zombie. */
603 1.107 thorpej p->p_stat = SZOMB;
604 1.107 thorpej
605 1.107 thorpej /* Wake up the parent so it can get exit status. */
606 1.107 thorpej if ((p->p_flag & P_FSTRACE) == 0 && p->p_exitsig != 0)
607 1.107 thorpej psignal(p->p_pptr, P_EXITSIG(p));
608 1.107 thorpej KERNEL_PROC_UNLOCK(curlwp);
609 1.107 thorpej wakeup((caddr_t)p->p_pptr);
610 1.107 thorpej }
611 1.56 thorpej }
612 1.56 thorpej }
613 1.56 thorpej
614 1.24 cgd int
615 1.107 thorpej sys_wait4(struct lwp *l, void *v, register_t *retval)
616 1.31 thorpej {
617 1.76 augustss struct sys_wait4_args /* {
618 1.89 lukem syscallarg(int) pid;
619 1.89 lukem syscallarg(int *) status;
620 1.89 lukem syscallarg(int) options;
621 1.89 lukem syscallarg(struct rusage *) rusage;
622 1.31 thorpej } */ *uap = v;
623 1.109 dsl struct proc *child, *parent;
624 1.109 dsl int status, error;
625 1.24 cgd
626 1.109 dsl parent = l->l_proc;
627 1.107 thorpej
628 1.26 cgd if (SCARG(uap, pid) == 0)
629 1.109 dsl SCARG(uap, pid) = -parent->p_pgid;
630 1.109 dsl if (SCARG(uap, options) & ~(WUNTRACED|WNOHANG|WALTSIG|WALLSIG))
631 1.24 cgd return (EINVAL);
632 1.41 mikel
633 1.109 dsl error = find_stopped_child(parent, SCARG(uap,pid), SCARG(uap,options),
634 1.109 dsl &child);
635 1.109 dsl if (error != 0)
636 1.109 dsl return error;
637 1.109 dsl if (child == NULL) {
638 1.109 dsl *retval = 0;
639 1.109 dsl return 0;
640 1.109 dsl }
641 1.109 dsl
642 1.109 dsl retval[0] = child->p_pid;
643 1.109 dsl
644 1.109 dsl if (child->p_stat == SZOMB) {
645 1.109 dsl if (SCARG(uap, status)) {
646 1.109 dsl status = child->p_xstat; /* convert to int */
647 1.109 dsl error = copyout((caddr_t)&status,
648 1.109 dsl (caddr_t)SCARG(uap, status),
649 1.109 dsl sizeof(status));
650 1.109 dsl if (error)
651 1.109 dsl return (error);
652 1.109 dsl }
653 1.109 dsl if (SCARG(uap, rusage)) {
654 1.109 dsl error = copyout((caddr_t)child->p_ru,
655 1.109 dsl (caddr_t)SCARG(uap, rusage),
656 1.109 dsl sizeof(struct rusage));
657 1.109 dsl if (error)
658 1.109 dsl return (error);
659 1.109 dsl }
660 1.109 dsl
661 1.109 dsl proc_free(child);
662 1.109 dsl return 0;
663 1.109 dsl }
664 1.109 dsl
665 1.109 dsl /* child state must be SSTOP */
666 1.109 dsl if (SCARG(uap, status)) {
667 1.109 dsl status = W_STOPCODE(child->p_xstat);
668 1.109 dsl return copyout((caddr_t)&status,
669 1.109 dsl (caddr_t)SCARG(uap, status),
670 1.109 dsl sizeof(status));
671 1.109 dsl }
672 1.109 dsl return 0;
673 1.109 dsl }
674 1.109 dsl
675 1.109 dsl /*
676 1.109 dsl * Scan list of child processes for a child process that has stopped or
677 1.109 dsl * exited. Used by sys_wait4 and 'compat' equivalents.
678 1.109 dsl */
679 1.109 dsl int
680 1.109 dsl find_stopped_child(struct proc *parent, pid_t pid, int options,
681 1.109 dsl struct proc **child_p)
682 1.109 dsl {
683 1.109 dsl struct proc *child;
684 1.109 dsl int c_found, error;
685 1.68 thorpej
686 1.120 yamt for (;;) {
687 1.109 dsl c_found = 0;
688 1.109 dsl LIST_FOREACH(child, &parent->p_children, p_sibling) {
689 1.109 dsl if (pid != WAIT_ANY &&
690 1.109 dsl child->p_pid != pid &&
691 1.109 dsl child->p_pgid != -pid)
692 1.109 dsl continue;
693 1.109 dsl /*
694 1.109 dsl * Wait for processes with p_exitsig != SIGCHLD
695 1.109 dsl * processes only if WALTSIG is set; wait for
696 1.109 dsl * processes with p_exitsig == SIGCHLD only
697 1.109 dsl * if WALTSIG is clear.
698 1.109 dsl */
699 1.109 dsl if (((options & WALLSIG) == 0) &&
700 1.109 dsl (options & WALTSIG ? child->p_exitsig == SIGCHLD
701 1.109 dsl : P_EXITSIG(child) != SIGCHLD))
702 1.109 dsl continue;
703 1.109 dsl
704 1.109 dsl c_found = 1;
705 1.109 dsl if (child->p_stat == SZOMB &&
706 1.109 dsl (options & WNOZOMBIE) == 0) {
707 1.109 dsl *child_p = child;
708 1.109 dsl return 0;
709 1.24 cgd }
710 1.109 dsl
711 1.109 dsl if (child->p_stat == SSTOP &&
712 1.109 dsl (child->p_flag & P_WAITED) == 0 &&
713 1.109 dsl (child->p_flag & P_TRACED || options & WUNTRACED)) {
714 1.109 dsl if ((options & WNOWAIT) == 0)
715 1.109 dsl child->p_flag |= P_WAITED;
716 1.109 dsl *child_p = child;
717 1.109 dsl return 0;
718 1.24 cgd }
719 1.109 dsl }
720 1.109 dsl if (c_found == 0)
721 1.109 dsl return ECHILD;
722 1.109 dsl if (options & WNOHANG) {
723 1.109 dsl *child_p = NULL;
724 1.109 dsl return 0;
725 1.109 dsl }
726 1.109 dsl error = tsleep((caddr_t)parent, PWAIT | PCATCH, "wait", 0);
727 1.109 dsl if (error != 0)
728 1.109 dsl return error;
729 1.109 dsl }
730 1.109 dsl }
731 1.109 dsl
732 1.109 dsl /*
733 1.109 dsl * Free a process after parent has taken all the state info.
734 1.109 dsl */
735 1.109 dsl void
736 1.109 dsl proc_free(struct proc *p)
737 1.109 dsl {
738 1.109 dsl struct proc *parent = p->p_pptr;
739 1.109 dsl int s;
740 1.24 cgd
741 1.109 dsl /*
742 1.109 dsl * If we got the child via ptrace(2) or procfs, and
743 1.109 dsl * the parent is different (meaning the process was
744 1.109 dsl * attached, rather than run as a child), then we need
745 1.109 dsl * to give it back to the old parent, and send the
746 1.109 dsl * parent the exit signal. The rest of the cleanup
747 1.109 dsl * will be done when the old parent waits on the child.
748 1.109 dsl */
749 1.109 dsl if ((p->p_flag & P_TRACED) && p->p_opptr != parent){
750 1.109 dsl parent = p->p_opptr;
751 1.109 dsl if (parent == NULL)
752 1.109 dsl parent = initproc;
753 1.109 dsl proc_reparent(p, parent);
754 1.109 dsl p->p_opptr = NULL;
755 1.109 dsl p->p_flag &= ~(P_TRACED|P_WAITED|P_FSTRACE);
756 1.109 dsl if (p->p_exitsig != 0)
757 1.109 dsl psignal(parent, P_EXITSIG(p));
758 1.109 dsl wakeup((caddr_t)parent);
759 1.109 dsl return;
760 1.109 dsl }
761 1.109 dsl
762 1.109 dsl scheduler_wait_hook(parent, p);
763 1.109 dsl p->p_xstat = 0;
764 1.115 dsl
765 1.109 dsl ruadd(&parent->p_stats->p_cru, p->p_ru);
766 1.115 dsl
767 1.115 dsl /*
768 1.115 dsl * At this point we are going to start freeing the final resources.
769 1.115 dsl * If anyone tries to access the proc structure after here they
770 1.115 dsl * will get a shock - bits are missing.
771 1.115 dsl * Attempt to make it hard!
772 1.115 dsl */
773 1.115 dsl
774 1.115 dsl p->p_stat = SIDL; /* not even a zombie any more */
775 1.115 dsl
776 1.109 dsl pool_put(&rusage_pool, p->p_ru);
777 1.56 thorpej
778 1.109 dsl /*
779 1.109 dsl * Finally finished with old proc entry.
780 1.109 dsl * Unlink it from its process group and free it.
781 1.109 dsl */
782 1.109 dsl leavepgrp(p);
783 1.56 thorpej
784 1.109 dsl s = proclist_lock_write();
785 1.109 dsl LIST_REMOVE(p, p_list); /* off zombproc */
786 1.115 dsl LIST_REMOVE(p, p_sibling);
787 1.109 dsl proclist_unlock_write(s);
788 1.35 mycroft
789 1.109 dsl /*
790 1.109 dsl * Decrement the count of procs running with this uid.
791 1.109 dsl */
792 1.109 dsl (void)chgproccnt(p->p_cred->p_ruid, -1);
793 1.24 cgd
794 1.109 dsl /*
795 1.109 dsl * Free up credentials.
796 1.109 dsl */
797 1.109 dsl if (--p->p_cred->p_refcnt == 0) {
798 1.109 dsl crfree(p->p_cred->pc_ucred);
799 1.109 dsl pool_put(&pcred_pool, p->p_cred);
800 1.109 dsl }
801 1.107 thorpej
802 1.109 dsl /*
803 1.109 dsl * Release reference to text vnode
804 1.109 dsl */
805 1.109 dsl if (p->p_textvp)
806 1.109 dsl vrele(p->p_textvp);
807 1.24 cgd
808 1.109 dsl /*
809 1.109 dsl * Release any SA state
810 1.109 dsl */
811 1.109 dsl if (p->p_sa) {
812 1.109 dsl free(p->p_sa->sa_stacks, M_SA);
813 1.109 dsl pool_put(&sadata_pool, p->p_sa);
814 1.24 cgd }
815 1.109 dsl
816 1.115 dsl /* Free proc structure and let pid be reallocated */
817 1.115 dsl proc_free_mem(p);
818 1.24 cgd }
819 1.24 cgd
820 1.24 cgd /*
821 1.24 cgd * make process 'parent' the new parent of process 'child'.
822 1.24 cgd */
823 1.24 cgd void
824 1.82 thorpej proc_reparent(struct proc *child, struct proc *parent)
825 1.24 cgd {
826 1.24 cgd
827 1.24 cgd if (child->p_pptr == parent)
828 1.24 cgd return;
829 1.70 thorpej
830 1.70 thorpej if (parent == initproc)
831 1.70 thorpej child->p_exitsig = SIGCHLD;
832 1.24 cgd
833 1.25 mycroft LIST_REMOVE(child, p_sibling);
834 1.25 mycroft LIST_INSERT_HEAD(&parent->p_children, child, p_sibling);
835 1.24 cgd child->p_pptr = parent;
836 1.24 cgd }
837