xref: /src/sys/kern/kern_exit.c

/*	$NetBSD: kern_exit.c,v 1.45 1997/04/28 13:17:05 mycroft Exp $	*/

/*
 * Copyright (c) 1982, 1986, 1989, 1991, 1993
 *	The Regents of the University of California.  All rights reserved.
 * (c) UNIX System Laboratories, Inc.
 * All or some portions of this file are derived from material licensed
 * to the University of California by American Telephone and Telegraph
 * Co. or Unix System Laboratories, Inc. and are reproduced herein with
 * the permission of UNIX System Laboratories, Inc.
 *
 * Redistribution and use in source and binary forms, with or without
 * modification, are permitted provided that the following conditions
 * are met:
 * 1. Redistributions of source code must retain the above copyright
 *    notice, this list of conditions and the following disclaimer.
 * 2. Redistributions in binary form must reproduce the above copyright
 *    notice, this list of conditions and the following disclaimer in the
 *    documentation and/or other materials provided with the distribution.
 * 3. All advertising materials mentioning features or use of this software
 *    must display the following acknowledgement:
 *	This product includes software developed by the University of
 *	California, Berkeley and its contributors.
 * 4. Neither the name of the University nor the names of its contributors
 *    may be used to endorse or promote products derived from this software
 *    without specific prior written permission.
 *
 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
 * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
 * SUCH DAMAGE.
 *
 *	@(#)kern_exit.c	8.7 (Berkeley) 2/12/94
 */

#include <sys/param.h>
#include <sys/systm.h>
#include <sys/map.h>
#include <sys/ioctl.h>
#include <sys/proc.h>
#include <sys/tty.h>
#include <sys/time.h>
#include <sys/resource.h>
#include <sys/kernel.h>
#include <sys/proc.h>
#include <sys/buf.h>
#include <sys/wait.h>
#include <sys/file.h>
#include <sys/vnode.h>
#include <sys/syslog.h>
#include <sys/malloc.h>
#include <sys/resourcevar.h>
#include <sys/ptrace.h>
#include <sys/acct.h>
#include <sys/filedesc.h>
#include <sys/signalvar.h>
#ifdef SYSVSHM
#include <sys/shm.h>
#endif
#ifdef SYSVSEM
#include <sys/sem.h>
#endif

#include <sys/mount.h>
#include <sys/syscallargs.h>

#include <machine/cpu.h>

#include <vm/vm.h>
#include <vm/vm_kern.h>

/*
 * exit --
 *	Death of process.
 */
int
sys_exit(p, v, retval)
	struct proc *p;
	void *v;
	register_t *retval;
{
	struct sys_exit_args /* {
		syscallarg(int) rval;
	} */ *uap = v;

	exit1(p, W_EXITCODE(SCARG(uap, rval), 0));
	/* NOTREACHED */
	return (0);
}

/*
 * Exit: deallocate address space and other resources, change proc state
 * to zombie, and unlink proc from allproc and parent's lists.  Save exit
 * status and rusage for wait().  Check for child processes and orphan them.
 */
void
exit1(p, rv)
	register struct proc *p;
	int rv;
{
	register struct proc *q, *nq;
	register struct vmspace *vm;

	if (p->p_pid == 1)
		panic("init died (signal %d, exit %d)",
		    WTERMSIG(rv), WEXITSTATUS(rv));
#ifdef PGINPROF
	vmsizmon();
#endif
	if (p->p_flag & P_PROFIL)
		stopprofclock(p);
	MALLOC(p->p_ru, struct rusage *, sizeof(struct rusage),
		M_ZOMBIE, M_WAITOK);
	/*
	 * If parent is waiting for us to exit or exec, P_PPWAIT is set; we
	 * wake up the parent early to avoid deadlock.
	 */
	p->p_flag |= P_WEXIT;
	if (p->p_flag & P_PPWAIT) {
		p->p_flag &= ~P_PPWAIT;
		wakeup((caddr_t)p->p_pptr);
	}
	p->p_sigignore = ~0;
	p->p_siglist = 0;
	untimeout(realitexpire, (caddr_t)p);

	/*
	 * Close open files and release open-file table.
	 * This may block!
	 */
	fdfree(p);

	/* The next three chunks should probably be moved to vmspace_exit. */
	vm = p->p_vmspace;
#ifdef SYSVSHM
	if (vm->vm_shm)
		shmexit(p);
#endif
#ifdef SYSVSEM
	semexit(p);
#endif
	/*
	 * Release user portion of address space.
	 * This releases references to vnodes,
	 * which could cause I/O if the file has been unlinked.
	 * Need to do this early enough that we can still sleep.
	 * Can't free the entire vmspace as the kernel stack
	 * may be mapped within that space also.
	 */
	if (vm->vm_refcnt == 1)
		(void) vm_map_remove(&vm->vm_map, VM_MIN_ADDRESS,
		    VM_MAXUSER_ADDRESS);

	if (SESS_LEADER(p)) {
		register struct session *sp = p->p_session;

		if (sp->s_ttyvp) {
			/*
			 * Controlling process.
			 * Signal foreground pgrp,
			 * drain controlling terminal
			 * and revoke access to controlling terminal.
			 */
			if (sp->s_ttyp->t_session == sp) {
				if (sp->s_ttyp->t_pgrp)
					pgsignal(sp->s_ttyp->t_pgrp, SIGHUP, 1);
				(void) ttywait(sp->s_ttyp);
				/*
				 * The tty could have been revoked
				 * if we blocked.
				 */
				if (sp->s_ttyvp)
					vgoneall(sp->s_ttyvp);
			}
			if (sp->s_ttyvp)
				vrele(sp->s_ttyvp);
			sp->s_ttyvp = NULL;
			/*
			 * s_ttyp is not zero'd; we use this to indicate
			 * that the session once had a controlling terminal.
			 * (for logging and informational purposes)
			 */
		}
		sp->s_leader = NULL;
	}
	fixjobc(p, p->p_pgrp, 0);
	p->p_rlimit[RLIMIT_FSIZE].rlim_cur = RLIM_INFINITY;
	(void)acct_process(p);
#ifdef KTRACE
	/*
	 * release trace file
	 */
	p->p_traceflag = 0;	/* don't trace the vrele() */
	if (p->p_tracep)
		vrele(p->p_tracep);
#endif
	/*
	 * Remove proc from allproc queue and pidhash chain.
	 * Place onto zombproc.  Unlink from parent's child list.
	 */
	LIST_REMOVE(p, p_list);
	LIST_INSERT_HEAD(&zombproc, p, p_list);
	p->p_stat = SZOMB;

	LIST_REMOVE(p, p_hash);

	q = p->p_children.lh_first;
	if (q)		/* only need this if any child is S_ZOMB */
		wakeup((caddr_t)initproc);
	for (; q != 0; q = nq) {
		nq = q->p_sibling.le_next;
		proc_reparent(q, initproc);
		/*
		 * Traced processes are killed
		 * since their existence means someone is screwing up.
		 */
		if (q->p_flag & P_TRACED) {
			q->p_flag &= ~(P_TRACED|P_WAITED|P_FSTRACE);
			psignal(q, SIGKILL);
		}
	}

	/*
	 * Save exit status and final rusage info, adding in child rusage
	 * info and self times.
	 */
	p->p_xstat = rv;
	*p->p_ru = p->p_stats->p_ru;
	calcru(p, &p->p_ru->ru_utime, &p->p_ru->ru_stime, NULL);
	ruadd(p->p_ru, &p->p_stats->p_cru);

	/*
	 * Notify parent that we're gone.
	 */
	if ((p->p_flag & P_FSTRACE) == 0)
		psignal(p->p_pptr, SIGCHLD);
	wakeup((caddr_t)p->p_pptr);

	/*
	 * Clear curproc after we've done all operations
	 * that could block, and before tearing down the rest
	 * of the process state that might be used from clock, etc.
	 * Also, can't clear curproc while we're still runnable,
	 * as we're not on a run queue (we are current, just not
	 * a proper proc any longer!).
	 *
	 * Other substructures are freed from wait().
	 */
	curproc = NULL;
	if (--p->p_limit->p_refcnt == 0)
		FREE(p->p_limit, M_SUBPROC);

	/*
	 * Finally, call machine-dependent code to release the remaining
	 * resources including address space, the kernel stack and pcb.
	 * The address space is released by "vmspace_free(p->p_vmspace)";
	 * This is machine-dependent, as we may have to change stacks
	 * or ensure that the current one isn't reallocated before we
	 * finish.  cpu_exit will end with a call to cpu_swtch(), finishing
	 * our execution (pun intended).
	 */
	cpu_exit(p);
}

int
sys_wait4(q, v, retval)
	register struct proc *q;
	void *v;
	register_t *retval;
{
	register struct sys_wait4_args /* {
		syscallarg(int) pid;
		syscallarg(int *) status;
		syscallarg(int) options;
		syscallarg(struct rusage *) rusage;
	} */ *uap = v;
	register int nfound;
	register struct proc *p, *t;
	int status, error;

	if (SCARG(uap, pid) == 0)
		SCARG(uap, pid) = -q->p_pgid;
	if (SCARG(uap, options) &~ (WUNTRACED|WNOHANG))
		return (EINVAL);

loop:
	nfound = 0;
	for (p = q->p_children.lh_first; p != 0; p = p->p_sibling.le_next) {
		if (SCARG(uap, pid) != WAIT_ANY &&
		    p->p_pid != SCARG(uap, pid) &&
		    p->p_pgid != -SCARG(uap, pid))
			continue;
		nfound++;
		if (p->p_stat == SZOMB) {
			retval[0] = p->p_pid;

			if (SCARG(uap, status)) {
				status = p->p_xstat;	/* convert to int */
				error = copyout((caddr_t)&status,
						(caddr_t)SCARG(uap, status),
						sizeof(status));
				if (error)
					return (error);
			}
			if (SCARG(uap, rusage) &&
			    (error = copyout((caddr_t)p->p_ru,
			    (caddr_t)SCARG(uap, rusage),
			    sizeof (struct rusage))))
				return (error);
			/*
			 * If we got the child via ptrace(2) or procfs, and
			 * the parent is different (meaning the process was
			 * attached, rather than run as a child), then we need
			 * to give it back to the old parent, and send the
			 * parent a SIGCHLD.  The rest of the cleanup will be
			 * done when the old parent waits on the child.
			 */
			if ((p->p_flag & P_TRACED) &&
			    p->p_oppid != p->p_pptr->p_pid) {
				t = pfind(p->p_oppid);
				proc_reparent(p, t ? t : initproc);
				p->p_oppid = 0;
				p->p_flag &= ~(P_TRACED|P_WAITED|P_FSTRACE);
				psignal(p->p_pptr, SIGCHLD);
				wakeup((caddr_t)p->p_pptr);
				return (0);
			}
			p->p_xstat = 0;
			ruadd(&q->p_stats->p_cru, p->p_ru);
			FREE(p->p_ru, M_ZOMBIE);

			/*
			 * Finally finished with old proc entry.
			 * Unlink it from its process group and free it.
			 */
			leavepgrp(p);
			LIST_REMOVE(p, p_list);	/* off zombproc */
			LIST_REMOVE(p, p_sibling);

			/*
			 * Decrement the count of procs running with this uid.
			 */
			(void)chgproccnt(p->p_cred->p_ruid, -1);

			/*
			 * Free up credentials.
			 */
			if (--p->p_cred->p_refcnt == 0) {
				crfree(p->p_cred->pc_ucred);
				FREE(p->p_cred, M_SUBPROC);
			}

			/*
			 * Release reference to text vnode
			 */
			if (p->p_textvp)
				vrele(p->p_textvp);

			/*
			 * Give machine-dependent layer a chance
			 * to free anything that cpu_exit couldn't
			 * release while still running in process context.
			 */
			cpu_wait(p);
			FREE(p, M_PROC);
			nprocs--;
			return (0);
		}
		if (p->p_stat == SSTOP && (p->p_flag & P_WAITED) == 0 &&
		    (p->p_flag & P_TRACED || SCARG(uap, options) & WUNTRACED)) {
			p->p_flag |= P_WAITED;
			retval[0] = p->p_pid;

			if (SCARG(uap, status)) {
				status = W_STOPCODE(p->p_xstat);
				error = copyout((caddr_t)&status,
				    (caddr_t)SCARG(uap, status),
				    sizeof(status));
			} else
				error = 0;
			return (error);
		}
	}
	if (nfound == 0)
		return (ECHILD);
	if (SCARG(uap, options) & WNOHANG) {
		retval[0] = 0;
		return (0);
	}
	if ((error = tsleep((caddr_t)q, PWAIT | PCATCH, "wait", 0)) != 0)
		return (error);
	goto loop;
}

/*
 * make process 'parent' the new parent of process 'child'.
 */
void
proc_reparent(child, parent)
	register struct proc *child;
	register struct proc *parent;
{

	if (child->p_pptr == parent)
		return;

	LIST_REMOVE(child, p_sibling);
	LIST_INSERT_HEAD(&parent->p_children, child, p_sibling);
	child->p_pptr = parent;
}