Home | History | Annotate | Line # | Download | only in kern
kern_proc.c revision 1.94.4.14
      1  1.94.4.14        ad /*	$NetBSD: kern_proc.c,v 1.94.4.14 2007/02/05 13:16:49 ad Exp $	*/
      2       1.33   thorpej 
      3       1.33   thorpej /*-
      4   1.94.4.8        ad  * Copyright (c) 1999, 2006, 2007 The NetBSD Foundation, Inc.
      5       1.33   thorpej  * All rights reserved.
      6       1.33   thorpej  *
      7       1.33   thorpej  * This code is derived from software contributed to The NetBSD Foundation
      8       1.33   thorpej  * by Jason R. Thorpe of the Numerical Aerospace Simulation Facility,
      9   1.94.4.5        ad  * NASA Ames Research Center, and by Andrew Doran.
     10       1.33   thorpej  *
     11       1.33   thorpej  * Redistribution and use in source and binary forms, with or without
     12       1.33   thorpej  * modification, are permitted provided that the following conditions
     13       1.33   thorpej  * are met:
     14       1.33   thorpej  * 1. Redistributions of source code must retain the above copyright
     15       1.33   thorpej  *    notice, this list of conditions and the following disclaimer.
     16       1.33   thorpej  * 2. Redistributions in binary form must reproduce the above copyright
     17       1.33   thorpej  *    notice, this list of conditions and the following disclaimer in the
     18       1.33   thorpej  *    documentation and/or other materials provided with the distribution.
     19       1.33   thorpej  * 3. All advertising materials mentioning features or use of this software
     20       1.33   thorpej  *    must display the following acknowledgement:
     21       1.33   thorpej  *	This product includes software developed by the NetBSD
     22       1.33   thorpej  *	Foundation, Inc. and its contributors.
     23       1.33   thorpej  * 4. Neither the name of The NetBSD Foundation nor the names of its
     24       1.33   thorpej  *    contributors may be used to endorse or promote products derived
     25       1.33   thorpej  *    from this software without specific prior written permission.
     26       1.33   thorpej  *
     27       1.33   thorpej  * THIS SOFTWARE IS PROVIDED BY THE NETBSD FOUNDATION, INC. AND CONTRIBUTORS
     28       1.33   thorpej  * ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
     29       1.33   thorpej  * TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
     30       1.33   thorpej  * PURPOSE ARE DISCLAIMED.  IN NO EVENT SHALL THE FOUNDATION OR CONTRIBUTORS
     31       1.33   thorpej  * BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
     32       1.33   thorpej  * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
     33       1.33   thorpej  * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
     34       1.33   thorpej  * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
     35       1.33   thorpej  * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
     36       1.33   thorpej  * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
     37       1.33   thorpej  * POSSIBILITY OF SUCH DAMAGE.
     38       1.33   thorpej  */
     39        1.9       cgd 
     40        1.1       cgd /*
     41        1.7       cgd  * Copyright (c) 1982, 1986, 1989, 1991, 1993
     42        1.7       cgd  *	The Regents of the University of California.  All rights reserved.
     43        1.1       cgd  *
     44        1.1       cgd  * Redistribution and use in source and binary forms, with or without
     45        1.1       cgd  * modification, are permitted provided that the following conditions
     46        1.1       cgd  * are met:
     47        1.1       cgd  * 1. Redistributions of source code must retain the above copyright
     48        1.1       cgd  *    notice, this list of conditions and the following disclaimer.
     49        1.1       cgd  * 2. Redistributions in binary form must reproduce the above copyright
     50        1.1       cgd  *    notice, this list of conditions and the following disclaimer in the
     51        1.1       cgd  *    documentation and/or other materials provided with the distribution.
     52       1.65       agc  * 3. Neither the name of the University nor the names of its contributors
     53        1.1       cgd  *    may be used to endorse or promote products derived from this software
     54        1.1       cgd  *    without specific prior written permission.
     55        1.1       cgd  *
     56        1.1       cgd  * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
     57        1.1       cgd  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
     58        1.1       cgd  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
     59        1.1       cgd  * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
     60        1.1       cgd  * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
     61        1.1       cgd  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
     62        1.1       cgd  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
     63        1.1       cgd  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
     64        1.1       cgd  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
     65        1.1       cgd  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
     66        1.1       cgd  * SUCH DAMAGE.
     67        1.1       cgd  *
     68       1.23      fvdl  *	@(#)kern_proc.c	8.7 (Berkeley) 2/14/95
     69        1.1       cgd  */
     70       1.45     lukem 
     71       1.45     lukem #include <sys/cdefs.h>
     72  1.94.4.14        ad __KERNEL_RCSID(0, "$NetBSD: kern_proc.c,v 1.94.4.14 2007/02/05 13:16:49 ad Exp $");
     73       1.48      yamt 
     74       1.48      yamt #include "opt_kstack.h"
     75       1.88      onoe #include "opt_maxuprc.h"
     76       1.90       rjs #include "opt_multiprocessor.h"
     77       1.90       rjs #include "opt_lockdebug.h"
     78        1.1       cgd 
     79        1.5   mycroft #include <sys/param.h>
     80        1.5   mycroft #include <sys/systm.h>
     81        1.5   mycroft #include <sys/kernel.h>
     82        1.5   mycroft #include <sys/proc.h>
     83       1.28   thorpej #include <sys/resourcevar.h>
     84        1.5   mycroft #include <sys/buf.h>
     85        1.5   mycroft #include <sys/acct.h>
     86        1.5   mycroft #include <sys/wait.h>
     87        1.5   mycroft #include <sys/file.h>
     88        1.8   mycroft #include <ufs/ufs/quota.h>
     89        1.5   mycroft #include <sys/uio.h>
     90        1.5   mycroft #include <sys/malloc.h>
     91       1.24   thorpej #include <sys/pool.h>
     92        1.5   mycroft #include <sys/mbuf.h>
     93        1.5   mycroft #include <sys/ioctl.h>
     94        1.5   mycroft #include <sys/tty.h>
     95       1.11       cgd #include <sys/signalvar.h>
     96       1.51  gmcgarry #include <sys/ras.h>
     97       1.81  junyoung #include <sys/filedesc.h>
     98       1.89      elad #include <sys/kauth.h>
     99   1.94.4.3        ad #include <sys/sleepq.h>
    100       1.81  junyoung 
    101       1.81  junyoung #include <uvm/uvm.h>
    102       1.79      yamt #include <uvm/uvm_extern.h>
    103        1.5   mycroft 
    104        1.7       cgd /*
    105       1.10   mycroft  * Other process lists
    106        1.7       cgd  */
    107       1.31   thorpej 
    108       1.10   mycroft struct proclist allproc;
    109       1.32   thorpej struct proclist zombproc;	/* resources have been freed */
    110       1.32   thorpej 
    111       1.32   thorpej /*
    112   1.94.4.7        ad  * There are two locks on global process state.
    113       1.33   thorpej  *
    114   1.94.4.1        ad  * 1. proclist_lock is a reader/writer lock and is used when modifying or
    115   1.94.4.1        ad  * examining process state from a process context.  It protects our internal
    116   1.94.4.1        ad  * tables, all of the process lists, and a number of members of struct lwp
    117   1.94.4.1        ad  * and struct proc.
    118   1.94.4.1        ad 
    119   1.94.4.1        ad  * 2. proclist_mutex is used when allproc must be traversed from an
    120   1.94.4.1        ad  * interrupt context, or when we must signal processes from an interrupt
    121   1.94.4.1        ad  * context.  The proclist_lock should always be used in preference.
    122       1.33   thorpej  *
    123   1.94.4.7        ad  *	proclist_lock	proclist_mutex	structure
    124   1.94.4.7        ad  *	--------------- --------------- -----------------
    125   1.94.4.7        ad  *	x				zombproc
    126   1.94.4.7        ad  *	x		x		pid_table
    127   1.94.4.7        ad  *	x				proc::p_pptr
    128   1.94.4.7        ad  *	x				proc::p_sibling
    129   1.94.4.7        ad  *	x				proc::p_children
    130   1.94.4.7        ad  *	x		x		allproc
    131   1.94.4.7        ad  *	x		x		proc::p_pgrp
    132   1.94.4.7        ad  *	x		x		proc::p_pglist
    133   1.94.4.7        ad  *	x		x		proc::p_session
    134   1.94.4.7        ad  *	x		x		proc::p_list
    135   1.94.4.7        ad  *			x		alllwp
    136   1.94.4.7        ad  *			x		lwp::l_list
    137   1.94.4.3        ad  *
    138   1.94.4.3        ad  * The lock order for processes and LWPs is apporoximately as following:
    139   1.94.4.3        ad  *
    140   1.94.4.3        ad  * kernel_mutex
    141   1.94.4.3        ad  * -> proclist_lock
    142   1.94.4.3        ad  *    -> proclist_mutex
    143   1.94.4.5        ad  *	-> proc::p_mutex
    144   1.94.4.3        ad  *         -> proc::p_smutex
    145       1.33   thorpej  */
    146   1.94.4.1        ad krwlock_t	proclist_lock;
    147   1.94.4.1        ad kmutex_t	proclist_mutex;
    148       1.33   thorpej 
    149       1.33   thorpej /*
    150       1.72  junyoung  * pid to proc lookup is done by indexing the pid_table array.
    151       1.61       dsl  * Since pid numbers are only allocated when an empty slot
    152       1.61       dsl  * has been found, there is no need to search any lists ever.
    153       1.61       dsl  * (an orphaned pgrp will lock the slot, a session will lock
    154       1.61       dsl  * the pgrp with the same number.)
    155       1.61       dsl  * If the table is too small it is reallocated with twice the
    156       1.61       dsl  * previous size and the entries 'unzipped' into the two halves.
    157       1.61       dsl  * A linked list of free entries is passed through the pt_proc
    158       1.61       dsl  * field of 'free' items - set odd to be an invalid ptr.
    159       1.61       dsl  */
    160       1.61       dsl 
    161       1.61       dsl struct pid_table {
    162       1.61       dsl 	struct proc	*pt_proc;
    163       1.61       dsl 	struct pgrp	*pt_pgrp;
    164       1.72  junyoung };
    165       1.61       dsl #if 1	/* strongly typed cast - should be a noop */
    166       1.84     perry static inline uint p2u(struct proc *p) { return (uint)(uintptr_t)p; }
    167       1.61       dsl #else
    168       1.61       dsl #define p2u(p) ((uint)p)
    169       1.72  junyoung #endif
    170       1.61       dsl #define P_VALID(p) (!(p2u(p) & 1))
    171       1.61       dsl #define P_NEXT(p) (p2u(p) >> 1)
    172       1.61       dsl #define P_FREE(pid) ((struct proc *)(uintptr_t)((pid) << 1 | 1))
    173       1.61       dsl 
    174       1.61       dsl #define INITIAL_PID_TABLE_SIZE	(1 << 5)
    175       1.61       dsl static struct pid_table *pid_table;
    176       1.61       dsl static uint pid_tbl_mask = INITIAL_PID_TABLE_SIZE - 1;
    177       1.61       dsl static uint pid_alloc_lim;	/* max we allocate before growing table */
    178       1.61       dsl static uint pid_alloc_cnt;	/* number of allocated pids */
    179       1.61       dsl 
    180       1.61       dsl /* links through free slots - never empty! */
    181       1.61       dsl static uint next_free_pt, last_free_pt;
    182       1.61       dsl static pid_t pid_max = PID_MAX;		/* largest value we allocate */
    183       1.31   thorpej 
    184       1.81  junyoung /* Components of the first process -- never freed. */
    185       1.81  junyoung struct session session0;
    186       1.81  junyoung struct pgrp pgrp0;
    187       1.81  junyoung struct proc proc0;
    188  1.94.4.13        ad struct lwp lwp0 __aligned(MIN_LWP_ALIGNMENT);
    189       1.89      elad kauth_cred_t cred0;
    190       1.81  junyoung struct filedesc0 filedesc0;
    191       1.81  junyoung struct cwdinfo cwdi0;
    192       1.81  junyoung struct plimit limit0;
    193       1.81  junyoung struct pstats pstat0;
    194       1.81  junyoung struct vmspace vmspace0;
    195       1.81  junyoung struct sigacts sigacts0;
    196   1.94.4.1        ad struct turnstile turnstile0;
    197       1.81  junyoung 
    198       1.81  junyoung extern struct user *proc0paddr;
    199       1.81  junyoung 
    200       1.81  junyoung extern const struct emul emul_netbsd;	/* defined in kern_exec.c */
    201       1.81  junyoung 
    202       1.81  junyoung int nofile = NOFILE;
    203       1.81  junyoung int maxuprc = MAXUPRC;
    204       1.81  junyoung int cmask = CMASK;
    205       1.81  junyoung 
    206       1.77    simonb POOL_INIT(proc_pool, sizeof(struct proc), 0, 0, 0, "procpl",
    207       1.77    simonb     &pool_allocator_nointr);
    208       1.77    simonb POOL_INIT(pgrp_pool, sizeof(struct pgrp), 0, 0, 0, "pgrppl",
    209       1.77    simonb     &pool_allocator_nointr);
    210       1.77    simonb POOL_INIT(plimit_pool, sizeof(struct plimit), 0, 0, 0, "plimitpl",
    211       1.77    simonb     &pool_allocator_nointr);
    212       1.77    simonb POOL_INIT(pstats_pool, sizeof(struct pstats), 0, 0, 0, "pstatspl",
    213       1.77    simonb     &pool_allocator_nointr);
    214       1.77    simonb POOL_INIT(rusage_pool, sizeof(struct rusage), 0, 0, 0, "rusgepl",
    215       1.77    simonb     &pool_allocator_nointr);
    216       1.77    simonb POOL_INIT(session_pool, sizeof(struct session), 0, 0, 0, "sessionpl",
    217       1.77    simonb     &pool_allocator_nointr);
    218       1.57   thorpej 
    219       1.57   thorpej MALLOC_DEFINE(M_EMULDATA, "emuldata", "Per-process emulation data");
    220       1.57   thorpej MALLOC_DEFINE(M_PROC, "proc", "Proc structures");
    221       1.57   thorpej MALLOC_DEFINE(M_SUBPROC, "subproc", "Proc sub-structures");
    222       1.10   mycroft 
    223       1.31   thorpej /*
    224       1.31   thorpej  * The process list descriptors, used during pid allocation and
    225       1.31   thorpej  * by sysctl.  No locking on this data structure is needed since
    226       1.31   thorpej  * it is completely static.
    227       1.31   thorpej  */
    228       1.31   thorpej const struct proclist_desc proclists[] = {
    229       1.31   thorpej 	{ &allproc	},
    230       1.31   thorpej 	{ &zombproc	},
    231       1.31   thorpej 	{ NULL		},
    232       1.31   thorpej };
    233       1.31   thorpej 
    234       1.72  junyoung static void orphanpg(struct pgrp *);
    235       1.72  junyoung static void pg_delete(pid_t);
    236       1.13  christos 
    237   1.94.4.6        ad static specificdata_domain_t proc_specificdata_domain;
    238   1.94.4.6        ad 
    239       1.10   mycroft /*
    240       1.10   mycroft  * Initialize global process hashing structures.
    241       1.10   mycroft  */
    242       1.11       cgd void
    243       1.59       dsl procinit(void)
    244        1.7       cgd {
    245       1.31   thorpej 	const struct proclist_desc *pd;
    246       1.61       dsl 	int i;
    247       1.61       dsl #define	LINK_EMPTY ((PID_MAX + INITIAL_PID_TABLE_SIZE) & ~(INITIAL_PID_TABLE_SIZE - 1))
    248       1.31   thorpej 
    249       1.31   thorpej 	for (pd = proclists; pd->pd_list != NULL; pd++)
    250       1.31   thorpej 		LIST_INIT(pd->pd_list);
    251        1.7       cgd 
    252  1.94.4.14        ad 	/*
    253  1.94.4.14        ad 	 * XXX p_smutex can be IPL_VM except for audio drivers
    254  1.94.4.14        ad 	 * XXX proclist_lock must die
    255  1.94.4.14        ad 	 */
    256   1.94.4.1        ad 	rw_init(&proclist_lock);
    257  1.94.4.14        ad 	mutex_init(&proclist_mutex, MUTEX_SPIN, IPL_SCHED);
    258       1.33   thorpej 
    259       1.61       dsl 	pid_table = malloc(INITIAL_PID_TABLE_SIZE * sizeof *pid_table,
    260       1.61       dsl 			    M_PROC, M_WAITOK);
    261       1.61       dsl 	/* Set free list running through table...
    262       1.61       dsl 	   Preset 'use count' above PID_MAX so we allocate pid 1 next. */
    263       1.61       dsl 	for (i = 0; i <= pid_tbl_mask; i++) {
    264       1.61       dsl 		pid_table[i].pt_proc = P_FREE(LINK_EMPTY + i + 1);
    265       1.61       dsl 		pid_table[i].pt_pgrp = 0;
    266       1.61       dsl 	}
    267       1.61       dsl 	/* slot 0 is just grabbed */
    268       1.61       dsl 	next_free_pt = 1;
    269       1.61       dsl 	/* Need to fix last entry. */
    270       1.61       dsl 	last_free_pt = pid_tbl_mask;
    271       1.61       dsl 	pid_table[last_free_pt].pt_proc = P_FREE(LINK_EMPTY);
    272       1.61       dsl 	/* point at which we grow table - to avoid reusing pids too often */
    273       1.61       dsl 	pid_alloc_lim = pid_tbl_mask - 1;
    274       1.61       dsl #undef LINK_EMPTY
    275       1.61       dsl 
    276       1.55   thorpej 	LIST_INIT(&alllwp);
    277       1.55   thorpej 
    278       1.43        ad 	uihashtbl =
    279       1.43        ad 	    hashinit(maxproc / 16, HASH_LIST, M_PROC, M_WAITOK, &uihash);
    280   1.94.4.6        ad 
    281   1.94.4.6        ad 	proc_specificdata_domain = specificdata_domain_create();
    282   1.94.4.6        ad 	KASSERT(proc_specificdata_domain != NULL);
    283        1.7       cgd }
    284        1.1       cgd 
    285        1.7       cgd /*
    286       1.81  junyoung  * Initialize process 0.
    287       1.81  junyoung  */
    288       1.81  junyoung void
    289       1.81  junyoung proc0_init(void)
    290       1.81  junyoung {
    291       1.81  junyoung 	struct proc *p;
    292       1.81  junyoung 	struct pgrp *pg;
    293       1.81  junyoung 	struct session *sess;
    294       1.81  junyoung 	struct lwp *l;
    295       1.81  junyoung 	u_int i;
    296       1.81  junyoung 	rlim_t lim;
    297       1.81  junyoung 
    298       1.81  junyoung 	p = &proc0;
    299       1.81  junyoung 	pg = &pgrp0;
    300       1.81  junyoung 	sess = &session0;
    301       1.81  junyoung 	l = &lwp0;
    302       1.81  junyoung 
    303  1.94.4.14        ad 	/* XXX p_smutex can be IPL_VM except for audio drivers */
    304  1.94.4.14        ad 	mutex_init(&p->p_smutex, MUTEX_SPIN, IPL_SCHED);
    305   1.94.4.7        ad 	mutex_init(&p->p_stmutex, MUTEX_SPIN, IPL_STATCLOCK);
    306   1.94.4.5        ad 	mutex_init(&p->p_rasmutex, MUTEX_SPIN, IPL_NONE);
    307   1.94.4.3        ad 	mutex_init(&p->p_mutex, MUTEX_DEFAULT, IPL_NONE);
    308   1.94.4.5        ad 	cv_init(&p->p_refcv, "drainref");
    309   1.94.4.5        ad 	cv_init(&p->p_waitcv, "wait");
    310   1.94.4.7        ad 	cv_init(&p->p_lwpcv, "lwpwait");
    311   1.94.4.5        ad 
    312       1.81  junyoung 	LIST_INIT(&p->p_lwps);
    313   1.94.4.3        ad 	LIST_INIT(&p->p_sigwaiters);
    314       1.81  junyoung 	LIST_INSERT_HEAD(&p->p_lwps, l, l_sibling);
    315   1.94.4.5        ad 
    316       1.81  junyoung 	p->p_nlwps = 1;
    317   1.94.4.3        ad 	p->p_nrlwps = 1;
    318   1.94.4.5        ad 	p->p_refcnt = 1;
    319       1.81  junyoung 
    320       1.81  junyoung 	pid_table[0].pt_proc = p;
    321       1.81  junyoung 	LIST_INSERT_HEAD(&allproc, p, p_list);
    322       1.81  junyoung 	LIST_INSERT_HEAD(&alllwp, l, l_list);
    323       1.81  junyoung 
    324       1.81  junyoung 	p->p_pgrp = pg;
    325       1.81  junyoung 	pid_table[0].pt_pgrp = pg;
    326       1.81  junyoung 	LIST_INIT(&pg->pg_members);
    327       1.81  junyoung 	LIST_INSERT_HEAD(&pg->pg_members, p, p_pglist);
    328       1.81  junyoung 
    329       1.81  junyoung 	pg->pg_session = sess;
    330       1.81  junyoung 	sess->s_count = 1;
    331       1.81  junyoung 	sess->s_sid = 0;
    332       1.81  junyoung 	sess->s_leader = p;
    333       1.81  junyoung 
    334       1.81  junyoung 	/*
    335       1.81  junyoung 	 * Set P_NOCLDWAIT so that kernel threads are reparented to
    336       1.81  junyoung 	 * init(8) when they exit.  init(8) can easily wait them out
    337       1.81  junyoung 	 * for us.
    338       1.81  junyoung 	 */
    339       1.81  junyoung 	p->p_flag = P_SYSTEM | P_NOCLDWAIT;
    340       1.81  junyoung 	p->p_stat = SACTIVE;
    341       1.81  junyoung 	p->p_nice = NZERO;
    342       1.81  junyoung 	p->p_emul = &emul_netbsd;
    343       1.81  junyoung #ifdef __HAVE_SYSCALL_INTERN
    344       1.81  junyoung 	(*p->p_emul->e_syscall_intern)(p);
    345       1.81  junyoung #endif
    346       1.81  junyoung 	strncpy(p->p_comm, "swapper", MAXCOMLEN);
    347       1.81  junyoung 
    348   1.94.4.4        ad 	l->l_mutex = &sched_mutex;
    349   1.94.4.3        ad 	l->l_flag = L_INMEM | L_SYSTEM;
    350       1.81  junyoung 	l->l_stat = LSONPROC;
    351   1.94.4.1        ad 	l->l_ts = &turnstile0;
    352   1.94.4.5        ad 	l->l_syncobj = &sched_syncobj;
    353   1.94.4.5        ad 	l->l_refcnt = 1;
    354   1.94.4.7        ad 	l->l_cpu = curcpu();
    355   1.94.4.8        ad 	l->l_priority = PRIBIO;
    356   1.94.4.8        ad 	l->l_usrpri = PRIBIO;
    357       1.81  junyoung 
    358       1.81  junyoung 	callout_init(&l->l_tsleep_ch);
    359   1.94.4.8        ad 	cv_init(&l->l_sigcv, "sigwait");
    360       1.81  junyoung 
    361       1.81  junyoung 	/* Create credentials. */
    362       1.89      elad 	cred0 = kauth_cred_alloc();
    363       1.89      elad 	p->p_cred = cred0;
    364   1.94.4.1        ad 	kauth_cred_hold(cred0);
    365   1.94.4.1        ad 	l->l_cred = cred0;
    366       1.81  junyoung 
    367       1.81  junyoung 	/* Create the CWD info. */
    368       1.81  junyoung 	p->p_cwdi = &cwdi0;
    369       1.81  junyoung 	cwdi0.cwdi_cmask = cmask;
    370       1.81  junyoung 	cwdi0.cwdi_refcnt = 1;
    371       1.81  junyoung 	simple_lock_init(&cwdi0.cwdi_slock);
    372       1.81  junyoung 
    373       1.81  junyoung 	/* Create the limits structures. */
    374       1.81  junyoung 	p->p_limit = &limit0;
    375       1.81  junyoung 	simple_lock_init(&limit0.p_slock);
    376       1.81  junyoung 	for (i = 0; i < sizeof(p->p_rlimit)/sizeof(p->p_rlimit[0]); i++)
    377       1.81  junyoung 		limit0.pl_rlimit[i].rlim_cur =
    378       1.81  junyoung 		    limit0.pl_rlimit[i].rlim_max = RLIM_INFINITY;
    379       1.81  junyoung 
    380       1.81  junyoung 	limit0.pl_rlimit[RLIMIT_NOFILE].rlim_max = maxfiles;
    381       1.81  junyoung 	limit0.pl_rlimit[RLIMIT_NOFILE].rlim_cur =
    382       1.81  junyoung 	    maxfiles < nofile ? maxfiles : nofile;
    383       1.81  junyoung 
    384       1.81  junyoung 	limit0.pl_rlimit[RLIMIT_NPROC].rlim_max = maxproc;
    385       1.81  junyoung 	limit0.pl_rlimit[RLIMIT_NPROC].rlim_cur =
    386       1.81  junyoung 	    maxproc < maxuprc ? maxproc : maxuprc;
    387       1.81  junyoung 
    388       1.81  junyoung 	lim = ptoa(uvmexp.free);
    389       1.81  junyoung 	limit0.pl_rlimit[RLIMIT_RSS].rlim_max = lim;
    390       1.81  junyoung 	limit0.pl_rlimit[RLIMIT_MEMLOCK].rlim_max = lim;
    391       1.81  junyoung 	limit0.pl_rlimit[RLIMIT_MEMLOCK].rlim_cur = lim / 3;
    392       1.81  junyoung 	limit0.pl_corename = defcorename;
    393       1.81  junyoung 	limit0.p_refcnt = 1;
    394       1.81  junyoung 
    395       1.81  junyoung 	/* Configure virtual memory system, set vm rlimits. */
    396       1.81  junyoung 	uvm_init_limits(p);
    397       1.81  junyoung 
    398       1.81  junyoung 	/* Initialize file descriptor table for proc0. */
    399       1.81  junyoung 	p->p_fd = &filedesc0.fd_fd;
    400       1.81  junyoung 	fdinit1(&filedesc0);
    401       1.81  junyoung 
    402       1.81  junyoung 	/*
    403       1.81  junyoung 	 * Initialize proc0's vmspace, which uses the kernel pmap.
    404       1.81  junyoung 	 * All kernel processes (which never have user space mappings)
    405       1.81  junyoung 	 * share proc0's vmspace, and thus, the kernel pmap.
    406       1.81  junyoung 	 */
    407       1.81  junyoung 	uvmspace_init(&vmspace0, pmap_kernel(), round_page(VM_MIN_ADDRESS),
    408       1.81  junyoung 	    trunc_page(VM_MAX_ADDRESS));
    409       1.81  junyoung 	p->p_vmspace = &vmspace0;
    410       1.81  junyoung 
    411       1.81  junyoung 	l->l_addr = proc0paddr;				/* XXX */
    412       1.81  junyoung 
    413       1.81  junyoung 	p->p_stats = &pstat0;
    414       1.81  junyoung 
    415       1.81  junyoung 	/* Initialize signal state for proc0. */
    416       1.81  junyoung 	p->p_sigacts = &sigacts0;
    417   1.94.4.5        ad 	mutex_init(&p->p_sigacts->sa_mutex, MUTEX_SPIN, IPL_NONE);
    418       1.81  junyoung 	siginit(p);
    419   1.94.4.6        ad 
    420   1.94.4.6        ad 	proc_initspecific(p);
    421   1.94.4.6        ad 	lwp_initspecific(l);
    422       1.81  junyoung }
    423       1.81  junyoung 
    424       1.81  junyoung /*
    425       1.74  junyoung  * Check that the specified process group is in the session of the
    426       1.60       dsl  * specified process.
    427       1.60       dsl  * Treats -ve ids as process ids.
    428       1.60       dsl  * Used to validate TIOCSPGRP requests.
    429       1.60       dsl  */
    430       1.60       dsl int
    431       1.60       dsl pgid_in_session(struct proc *p, pid_t pg_id)
    432       1.60       dsl {
    433       1.60       dsl 	struct pgrp *pgrp;
    434       1.60       dsl 
    435       1.60       dsl 	if (pg_id < 0) {
    436       1.60       dsl 		struct proc *p1 = pfind(-pg_id);
    437       1.64       dsl 		if (p1 == NULL)
    438       1.64       dsl 			return EINVAL;
    439       1.60       dsl 		pgrp = p1->p_pgrp;
    440       1.60       dsl 	} else {
    441       1.60       dsl 		pgrp = pgfind(pg_id);
    442       1.60       dsl 		if (pgrp == NULL)
    443       1.64       dsl 			return EINVAL;
    444       1.60       dsl 	}
    445       1.60       dsl 	if (pgrp->pg_session != p->p_pgrp->pg_session)
    446       1.60       dsl 		return EPERM;
    447       1.60       dsl 	return 0;
    448        1.7       cgd }
    449        1.4    andrew 
    450        1.1       cgd /*
    451       1.41  sommerfe  * Is p an inferior of q?
    452       1.94        ad  *
    453       1.94        ad  * Call with the proclist_lock held.
    454        1.1       cgd  */
    455       1.11       cgd int
    456       1.59       dsl inferior(struct proc *p, struct proc *q)
    457        1.1       cgd {
    458        1.1       cgd 
    459       1.41  sommerfe 	for (; p != q; p = p->p_pptr)
    460        1.1       cgd 		if (p->p_pid == 0)
    461       1.82  junyoung 			return 0;
    462       1.82  junyoung 	return 1;
    463        1.1       cgd }
    464        1.1       cgd 
    465        1.1       cgd /*
    466        1.1       cgd  * Locate a process by number
    467        1.1       cgd  */
    468        1.1       cgd struct proc *
    469       1.68       dsl p_find(pid_t pid, uint flags)
    470        1.1       cgd {
    471       1.33   thorpej 	struct proc *p;
    472       1.68       dsl 	char stat;
    473        1.1       cgd 
    474       1.68       dsl 	if (!(flags & PFIND_LOCKED))
    475   1.94.4.1        ad 		rw_enter(&proclist_lock, RW_READER);
    476   1.94.4.1        ad 
    477       1.61       dsl 	p = pid_table[pid & pid_tbl_mask].pt_proc;
    478   1.94.4.7        ad 
    479       1.61       dsl 	/* Only allow live processes to be found by pid. */
    480   1.94.4.7        ad 	/* XXXSMP p_stat */
    481   1.94.4.7        ad 	if (P_VALID(p) && p->p_pid == pid && ((stat = p->p_stat) == SACTIVE ||
    482   1.94.4.7        ad 	    stat == SSTOP || ((flags & PFIND_ZOMBIE) &&
    483   1.94.4.7        ad 	    (stat == SZOMB || stat == SDEAD || stat == SDYING)))) {
    484       1.68       dsl 		if (flags & PFIND_UNLOCK_OK)
    485   1.94.4.1        ad 			 rw_exit(&proclist_lock);
    486       1.68       dsl 		return p;
    487       1.68       dsl 	}
    488       1.68       dsl 	if (flags & PFIND_UNLOCK_FAIL)
    489   1.94.4.1        ad 		 rw_exit(&proclist_lock);
    490       1.68       dsl 	return NULL;
    491        1.1       cgd }
    492        1.1       cgd 
    493       1.61       dsl 
    494        1.1       cgd /*
    495        1.1       cgd  * Locate a process group by number
    496        1.1       cgd  */
    497        1.1       cgd struct pgrp *
    498       1.68       dsl pg_find(pid_t pgid, uint flags)
    499        1.1       cgd {
    500       1.68       dsl 	struct pgrp *pg;
    501        1.1       cgd 
    502       1.68       dsl 	if (!(flags & PFIND_LOCKED))
    503   1.94.4.1        ad 		rw_enter(&proclist_lock, RW_READER);
    504       1.68       dsl 	pg = pid_table[pgid & pid_tbl_mask].pt_pgrp;
    505       1.61       dsl 	/*
    506       1.61       dsl 	 * Can't look up a pgrp that only exists because the session
    507       1.61       dsl 	 * hasn't died yet (traditional)
    508       1.61       dsl 	 */
    509       1.68       dsl 	if (pg == NULL || pg->pg_id != pgid || LIST_EMPTY(&pg->pg_members)) {
    510       1.68       dsl 		if (flags & PFIND_UNLOCK_FAIL)
    511   1.94.4.1        ad 			 rw_exit(&proclist_lock);
    512       1.68       dsl 		return NULL;
    513       1.68       dsl 	}
    514       1.68       dsl 
    515       1.68       dsl 	if (flags & PFIND_UNLOCK_OK)
    516   1.94.4.1        ad 		rw_exit(&proclist_lock);
    517       1.68       dsl 	return pg;
    518        1.1       cgd }
    519        1.1       cgd 
    520       1.61       dsl static void
    521       1.61       dsl expand_pid_table(void)
    522        1.1       cgd {
    523       1.61       dsl 	uint pt_size = pid_tbl_mask + 1;
    524       1.61       dsl 	struct pid_table *n_pt, *new_pt;
    525       1.61       dsl 	struct proc *proc;
    526       1.61       dsl 	struct pgrp *pgrp;
    527       1.61       dsl 	int i;
    528       1.61       dsl 	pid_t pid;
    529        1.1       cgd 
    530       1.61       dsl 	new_pt = malloc(pt_size * 2 * sizeof *new_pt, M_PROC, M_WAITOK);
    531       1.61       dsl 
    532   1.94.4.1        ad 	rw_enter(&proclist_lock, RW_WRITER);
    533       1.61       dsl 	if (pt_size != pid_tbl_mask + 1) {
    534       1.61       dsl 		/* Another process beat us to it... */
    535   1.94.4.1        ad 		rw_exit(&proclist_lock);
    536       1.61       dsl 		FREE(new_pt, M_PROC);
    537       1.61       dsl 		return;
    538       1.61       dsl 	}
    539       1.72  junyoung 
    540       1.61       dsl 	/*
    541       1.61       dsl 	 * Copy entries from old table into new one.
    542       1.61       dsl 	 * If 'pid' is 'odd' we need to place in the upper half,
    543       1.61       dsl 	 * even pid's to the lower half.
    544       1.61       dsl 	 * Free items stay in the low half so we don't have to
    545       1.61       dsl 	 * fixup the reference to them.
    546       1.61       dsl 	 * We stuff free items on the front of the freelist
    547       1.61       dsl 	 * because we can't write to unmodified entries.
    548       1.74  junyoung 	 * Processing the table backwards maintains a semblance
    549       1.61       dsl 	 * of issueing pid numbers that increase with time.
    550       1.61       dsl 	 */
    551       1.61       dsl 	i = pt_size - 1;
    552       1.61       dsl 	n_pt = new_pt + i;
    553       1.61       dsl 	for (; ; i--, n_pt--) {
    554       1.61       dsl 		proc = pid_table[i].pt_proc;
    555       1.61       dsl 		pgrp = pid_table[i].pt_pgrp;
    556       1.61       dsl 		if (!P_VALID(proc)) {
    557       1.61       dsl 			/* Up 'use count' so that link is valid */
    558       1.61       dsl 			pid = (P_NEXT(proc) + pt_size) & ~pt_size;
    559       1.61       dsl 			proc = P_FREE(pid);
    560       1.61       dsl 			if (pgrp)
    561       1.61       dsl 				pid = pgrp->pg_id;
    562       1.61       dsl 		} else
    563       1.61       dsl 			pid = proc->p_pid;
    564       1.72  junyoung 
    565       1.61       dsl 		/* Save entry in appropriate half of table */
    566       1.61       dsl 		n_pt[pid & pt_size].pt_proc = proc;
    567       1.61       dsl 		n_pt[pid & pt_size].pt_pgrp = pgrp;
    568       1.61       dsl 
    569       1.61       dsl 		/* Put other piece on start of free list */
    570       1.61       dsl 		pid = (pid ^ pt_size) & ~pid_tbl_mask;
    571       1.61       dsl 		n_pt[pid & pt_size].pt_proc =
    572       1.61       dsl 				    P_FREE((pid & ~pt_size) | next_free_pt);
    573       1.61       dsl 		n_pt[pid & pt_size].pt_pgrp = 0;
    574       1.61       dsl 		next_free_pt = i | (pid & pt_size);
    575       1.61       dsl 		if (i == 0)
    576       1.61       dsl 			break;
    577       1.61       dsl 	}
    578       1.61       dsl 
    579       1.61       dsl 	/* Switch tables */
    580   1.94.4.2        ad 	mutex_enter(&proclist_mutex);
    581       1.61       dsl 	n_pt = pid_table;
    582       1.61       dsl 	pid_table = new_pt;
    583   1.94.4.2        ad 	mutex_exit(&proclist_mutex);
    584       1.61       dsl 	pid_tbl_mask = pt_size * 2 - 1;
    585       1.61       dsl 
    586       1.61       dsl 	/*
    587       1.61       dsl 	 * pid_max starts as PID_MAX (= 30000), once we have 16384
    588       1.61       dsl 	 * allocated pids we need it to be larger!
    589       1.61       dsl 	 */
    590       1.61       dsl 	if (pid_tbl_mask > PID_MAX) {
    591       1.61       dsl 		pid_max = pid_tbl_mask * 2 + 1;
    592       1.61       dsl 		pid_alloc_lim |= pid_alloc_lim << 1;
    593       1.61       dsl 	} else
    594       1.61       dsl 		pid_alloc_lim <<= 1;	/* doubles number of free slots... */
    595       1.61       dsl 
    596   1.94.4.1        ad 	rw_exit(&proclist_lock);
    597       1.61       dsl 	FREE(n_pt, M_PROC);
    598       1.61       dsl }
    599       1.61       dsl 
    600       1.61       dsl struct proc *
    601       1.61       dsl proc_alloc(void)
    602       1.61       dsl {
    603       1.61       dsl 	struct proc *p;
    604       1.61       dsl 	int nxt;
    605       1.61       dsl 	pid_t pid;
    606       1.61       dsl 	struct pid_table *pt;
    607       1.61       dsl 
    608       1.61       dsl 	p = pool_get(&proc_pool, PR_WAITOK);
    609       1.61       dsl 	p->p_stat = SIDL;			/* protect against others */
    610       1.61       dsl 
    611   1.94.4.6        ad 	proc_initspecific(p);
    612       1.61       dsl 	/* allocate next free pid */
    613       1.61       dsl 
    614       1.61       dsl 	for (;;expand_pid_table()) {
    615       1.61       dsl 		if (__predict_false(pid_alloc_cnt >= pid_alloc_lim))
    616       1.61       dsl 			/* ensure pids cycle through 2000+ values */
    617       1.61       dsl 			continue;
    618   1.94.4.1        ad 		rw_enter(&proclist_lock, RW_WRITER);
    619       1.61       dsl 		pt = &pid_table[next_free_pt];
    620        1.1       cgd #ifdef DIAGNOSTIC
    621       1.63  christos 		if (__predict_false(P_VALID(pt->pt_proc) || pt->pt_pgrp))
    622       1.61       dsl 			panic("proc_alloc: slot busy");
    623        1.1       cgd #endif
    624       1.61       dsl 		nxt = P_NEXT(pt->pt_proc);
    625       1.61       dsl 		if (nxt & pid_tbl_mask)
    626       1.61       dsl 			break;
    627       1.61       dsl 		/* Table full - expand (NB last entry not used....) */
    628   1.94.4.1        ad 		rw_exit(&proclist_lock);
    629       1.61       dsl 	}
    630       1.61       dsl 
    631       1.61       dsl 	/* pid is 'saved use count' + 'size' + entry */
    632       1.61       dsl 	pid = (nxt & ~pid_tbl_mask) + pid_tbl_mask + 1 + next_free_pt;
    633       1.61       dsl 	if ((uint)pid > (uint)pid_max)
    634       1.61       dsl 		pid &= pid_tbl_mask;
    635       1.61       dsl 	p->p_pid = pid;
    636       1.61       dsl 	next_free_pt = nxt & pid_tbl_mask;
    637       1.61       dsl 
    638       1.61       dsl 	/* Grab table slot */
    639   1.94.4.2        ad 	mutex_enter(&proclist_mutex);
    640       1.61       dsl 	pt->pt_proc = p;
    641   1.94.4.2        ad 	mutex_exit(&proclist_mutex);
    642       1.61       dsl 	pid_alloc_cnt++;
    643       1.61       dsl 
    644   1.94.4.1        ad 	rw_exit(&proclist_lock);
    645       1.61       dsl 
    646       1.61       dsl 	return p;
    647       1.61       dsl }
    648       1.61       dsl 
    649       1.61       dsl /*
    650       1.61       dsl  * Free last resources of a process - called from proc_free (in kern_exit.c)
    651   1.94.4.1        ad  *
    652   1.94.4.1        ad  * Called with the proclist_lock write held, and releases upon exit.
    653       1.61       dsl  */
    654       1.61       dsl void
    655       1.61       dsl proc_free_mem(struct proc *p)
    656       1.61       dsl {
    657       1.61       dsl 	pid_t pid = p->p_pid;
    658       1.61       dsl 	struct pid_table *pt;
    659       1.61       dsl 
    660   1.94.4.1        ad 	LOCK_ASSERT(rw_write_held(&proclist_lock));
    661       1.61       dsl 
    662       1.61       dsl 	pt = &pid_table[pid & pid_tbl_mask];
    663        1.1       cgd #ifdef DIAGNOSTIC
    664       1.63  christos 	if (__predict_false(pt->pt_proc != p))
    665       1.61       dsl 		panic("proc_free: pid_table mismatch, pid %x, proc %p",
    666       1.61       dsl 			pid, p);
    667        1.1       cgd #endif
    668   1.94.4.2        ad 	mutex_enter(&proclist_mutex);
    669       1.61       dsl 	/* save pid use count in slot */
    670       1.61       dsl 	pt->pt_proc = P_FREE(pid & ~pid_tbl_mask);
    671       1.61       dsl 
    672       1.61       dsl 	if (pt->pt_pgrp == NULL) {
    673       1.61       dsl 		/* link last freed entry onto ours */
    674       1.61       dsl 		pid &= pid_tbl_mask;
    675       1.61       dsl 		pt = &pid_table[last_free_pt];
    676       1.61       dsl 		pt->pt_proc = P_FREE(P_NEXT(pt->pt_proc) | pid);
    677       1.61       dsl 		last_free_pt = pid;
    678       1.61       dsl 		pid_alloc_cnt--;
    679       1.61       dsl 	}
    680   1.94.4.2        ad 	mutex_exit(&proclist_mutex);
    681       1.61       dsl 
    682       1.61       dsl 	nprocs--;
    683   1.94.4.1        ad 	rw_exit(&proclist_lock);
    684       1.61       dsl 
    685       1.61       dsl 	pool_put(&proc_pool, p);
    686       1.61       dsl }
    687       1.61       dsl 
    688       1.61       dsl /*
    689       1.61       dsl  * Move p to a new or existing process group (and session)
    690       1.61       dsl  *
    691       1.61       dsl  * If we are creating a new pgrp, the pgid should equal
    692       1.72  junyoung  * the calling process' pid.
    693       1.61       dsl  * If is only valid to enter a process group that is in the session
    694       1.61       dsl  * of the process.
    695       1.61       dsl  * Also mksess should only be set if we are creating a process group
    696       1.61       dsl  *
    697       1.72  junyoung  * Only called from sys_setsid, sys_setpgid/sys_setpgrp and the
    698   1.94.4.1        ad  * SYSV setpgrp support for hpux.
    699       1.61       dsl  */
    700       1.61       dsl int
    701   1.94.4.1        ad enterpgrp(struct proc *curp, pid_t pid, pid_t pgid, int mksess)
    702       1.61       dsl {
    703       1.61       dsl 	struct pgrp *new_pgrp, *pgrp;
    704       1.61       dsl 	struct session *sess;
    705   1.94.4.1        ad 	struct proc *p;
    706       1.61       dsl 	int rval;
    707       1.61       dsl 	pid_t pg_id = NO_PGID;
    708       1.61       dsl 
    709       1.61       dsl 	/* Allocate data areas we might need before doing any validity checks */
    710   1.94.4.1        ad 	rw_enter(&proclist_lock, RW_READER);		/* Because pid_table might change */
    711       1.61       dsl 	if (pid_table[pgid & pid_tbl_mask].pt_pgrp == 0) {
    712   1.94.4.1        ad 		rw_exit(&proclist_lock);
    713       1.61       dsl 		new_pgrp = pool_get(&pgrp_pool, PR_WAITOK);
    714       1.61       dsl 	} else {
    715   1.94.4.1        ad 		rw_exit(&proclist_lock);
    716       1.61       dsl 		new_pgrp = NULL;
    717       1.61       dsl 	}
    718       1.61       dsl 	if (mksess)
    719   1.94.4.6        ad 		sess = pool_get(&session_pool, PR_WAITOK);
    720       1.61       dsl 	else
    721       1.61       dsl 		sess = NULL;
    722       1.61       dsl 
    723   1.94.4.1        ad 	rw_enter(&proclist_lock, RW_WRITER);
    724       1.61       dsl 	rval = EPERM;	/* most common error (to save typing) */
    725       1.61       dsl 
    726       1.61       dsl 	/* Check pgrp exists or can be created */
    727       1.61       dsl 	pgrp = pid_table[pgid & pid_tbl_mask].pt_pgrp;
    728       1.61       dsl 	if (pgrp != NULL && pgrp->pg_id != pgid)
    729       1.61       dsl 		goto done;
    730       1.61       dsl 
    731       1.61       dsl 	/* Can only set another process under restricted circumstances. */
    732   1.94.4.1        ad 	if (pid != curp->p_pid) {
    733       1.61       dsl 		/* must exist and be one of our children... */
    734   1.94.4.1        ad 		if ((p = p_find(pid, PFIND_LOCKED)) == NULL ||
    735   1.94.4.1        ad 		    !inferior(p, curp)) {
    736       1.61       dsl 			rval = ESRCH;
    737       1.61       dsl 			goto done;
    738       1.61       dsl 		}
    739       1.61       dsl 		/* ... in the same session... */
    740       1.61       dsl 		if (sess != NULL || p->p_session != curp->p_session)
    741       1.61       dsl 			goto done;
    742       1.61       dsl 		/* ... existing pgid must be in same session ... */
    743       1.61       dsl 		if (pgrp != NULL && pgrp->pg_session != p->p_session)
    744       1.61       dsl 			goto done;
    745       1.61       dsl 		/* ... and not done an exec. */
    746       1.61       dsl 		if (p->p_flag & P_EXEC) {
    747       1.61       dsl 			rval = EACCES;
    748       1.61       dsl 			goto done;
    749       1.49     enami 		}
    750   1.94.4.1        ad 	} else {
    751   1.94.4.1        ad 		/* ... setsid() cannot re-enter a pgrp */
    752   1.94.4.1        ad 		if (mksess && (curp->p_pgid == curp->p_pid ||
    753   1.94.4.1        ad 		    pg_find(curp->p_pid, PFIND_LOCKED)))
    754   1.94.4.1        ad 			goto done;
    755   1.94.4.1        ad 		p = curp;
    756       1.61       dsl 	}
    757        1.1       cgd 
    758       1.61       dsl 	/* Changing the process group/session of a session
    759       1.61       dsl 	   leader is definitely off limits. */
    760       1.61       dsl 	if (SESS_LEADER(p)) {
    761       1.61       dsl 		if (sess == NULL && p->p_pgrp == pgrp)
    762       1.61       dsl 			/* unless it's a definite noop */
    763       1.61       dsl 			rval = 0;
    764       1.61       dsl 		goto done;
    765       1.61       dsl 	}
    766       1.61       dsl 
    767       1.61       dsl 	/* Can only create a process group with id of process */
    768       1.61       dsl 	if (pgrp == NULL && pgid != pid)
    769       1.61       dsl 		goto done;
    770       1.61       dsl 
    771       1.61       dsl 	/* Can only create a session if creating pgrp */
    772       1.61       dsl 	if (sess != NULL && pgrp != NULL)
    773       1.61       dsl 		goto done;
    774       1.61       dsl 
    775       1.61       dsl 	/* Check we allocated memory for a pgrp... */
    776       1.61       dsl 	if (pgrp == NULL && new_pgrp == NULL)
    777       1.61       dsl 		goto done;
    778       1.61       dsl 
    779       1.61       dsl 	/* Don't attach to 'zombie' pgrp */
    780       1.61       dsl 	if (pgrp != NULL && LIST_EMPTY(&pgrp->pg_members))
    781       1.61       dsl 		goto done;
    782       1.61       dsl 
    783       1.61       dsl 	/* Expect to succeed now */
    784       1.61       dsl 	rval = 0;
    785       1.61       dsl 
    786       1.61       dsl 	if (pgrp == p->p_pgrp)
    787       1.61       dsl 		/* nothing to do */
    788       1.61       dsl 		goto done;
    789       1.61       dsl 
    790       1.61       dsl 	/* Ok all setup, link up required structures */
    791   1.94.4.2        ad 
    792       1.61       dsl 	if (pgrp == NULL) {
    793       1.61       dsl 		pgrp = new_pgrp;
    794       1.61       dsl 		new_pgrp = 0;
    795       1.61       dsl 		if (sess != NULL) {
    796       1.21   thorpej 			sess->s_sid = p->p_pid;
    797        1.1       cgd 			sess->s_leader = p;
    798        1.1       cgd 			sess->s_count = 1;
    799        1.1       cgd 			sess->s_ttyvp = NULL;
    800        1.1       cgd 			sess->s_ttyp = NULL;
    801       1.58       dsl 			sess->s_flags = p->p_session->s_flags & ~S_LOGIN_SET;
    802       1.25     perry 			memcpy(sess->s_login, p->p_session->s_login,
    803        1.1       cgd 			    sizeof(sess->s_login));
    804   1.94.4.5        ad 			p->p_lflag &= ~PL_CONTROLT;
    805        1.1       cgd 		} else {
    806       1.61       dsl 			sess = p->p_pgrp->pg_session;
    807       1.61       dsl 			SESSHOLD(sess);
    808        1.1       cgd 		}
    809       1.61       dsl 		pgrp->pg_session = sess;
    810       1.61       dsl 		sess = 0;
    811       1.61       dsl 
    812        1.1       cgd 		pgrp->pg_id = pgid;
    813       1.10   mycroft 		LIST_INIT(&pgrp->pg_members);
    814       1.61       dsl #ifdef DIAGNOSTIC
    815       1.63  christos 		if (__predict_false(pid_table[pgid & pid_tbl_mask].pt_pgrp))
    816       1.61       dsl 			panic("enterpgrp: pgrp table slot in use");
    817       1.63  christos 		if (__predict_false(mksess && p != curp))
    818       1.63  christos 			panic("enterpgrp: mksession and p != curproc");
    819       1.61       dsl #endif
    820   1.94.4.2        ad 		mutex_enter(&proclist_mutex);
    821       1.61       dsl 		pid_table[pgid & pid_tbl_mask].pt_pgrp = pgrp;
    822        1.1       cgd 		pgrp->pg_jobc = 0;
    823   1.94.4.2        ad 	} else
    824   1.94.4.2        ad 		mutex_enter(&proclist_mutex);
    825        1.1       cgd 
    826   1.94.4.1        ad #ifdef notyet
    827   1.94.4.1        ad 	/*
    828   1.94.4.1        ad 	 * If there's a controlling terminal for the current session, we
    829   1.94.4.1        ad 	 * have to interlock with it.  See ttread().
    830   1.94.4.1        ad 	 */
    831   1.94.4.1        ad 	if (p->p_session->s_ttyvp != NULL) {
    832   1.94.4.1        ad 		tp = p->p_session->s_ttyp;
    833   1.94.4.1        ad 		mutex_enter(&tp->t_mutex);
    834   1.94.4.1        ad 	} else
    835   1.94.4.1        ad 		tp = NULL;
    836   1.94.4.1        ad #endif
    837   1.94.4.1        ad 
    838        1.1       cgd 	/*
    839        1.1       cgd 	 * Adjust eligibility of affected pgrps to participate in job control.
    840        1.1       cgd 	 * Increment eligibility counts before decrementing, otherwise we
    841        1.1       cgd 	 * could reach 0 spuriously during the first call.
    842        1.1       cgd 	 */
    843        1.1       cgd 	fixjobc(p, pgrp, 1);
    844        1.1       cgd 	fixjobc(p, p->p_pgrp, 0);
    845        1.1       cgd 
    846   1.94.4.1        ad 	/* Move process to requested group. */
    847       1.10   mycroft 	LIST_REMOVE(p, p_pglist);
    848       1.52      matt 	if (LIST_EMPTY(&p->p_pgrp->pg_members))
    849       1.61       dsl 		/* defer delete until we've dumped the lock */
    850       1.61       dsl 		pg_id = p->p_pgrp->pg_id;
    851        1.1       cgd 	p->p_pgrp = pgrp;
    852       1.10   mycroft 	LIST_INSERT_HEAD(&pgrp->pg_members, p, p_pglist);
    853   1.94.4.1        ad 	mutex_exit(&proclist_mutex);
    854   1.94.4.1        ad 
    855   1.94.4.1        ad #ifdef notyet
    856   1.94.4.1        ad 	/* Done with the swap; we can release the tty mutex. */
    857   1.94.4.1        ad 	if (tp != NULL)
    858   1.94.4.1        ad 		mutex_exit(&tp->t_mutex);
    859   1.94.4.1        ad #endif
    860       1.61       dsl 
    861       1.61       dsl     done:
    862   1.94.4.5        ad 	if (pg_id != NO_PGID)
    863   1.94.4.5        ad 		pg_delete(pg_id);
    864   1.94.4.1        ad 	rw_exit(&proclist_lock);
    865       1.61       dsl 	if (sess != NULL)
    866       1.77    simonb 		pool_put(&session_pool, sess);
    867       1.61       dsl 	if (new_pgrp != NULL)
    868       1.61       dsl 		pool_put(&pgrp_pool, new_pgrp);
    869       1.63  christos #ifdef DEBUG_PGRP
    870       1.63  christos 	if (__predict_false(rval))
    871       1.61       dsl 		printf("enterpgrp(%d,%d,%d), curproc %d, rval %d\n",
    872       1.61       dsl 			pid, pgid, mksess, curp->p_pid, rval);
    873       1.61       dsl #endif
    874       1.61       dsl 	return rval;
    875        1.1       cgd }
    876        1.1       cgd 
    877        1.1       cgd /*
    878   1.94.4.5        ad  * Remove a process from its process group.  Must be called with the
    879   1.94.4.5        ad  * proclist_lock write held.
    880        1.1       cgd  */
    881   1.94.4.1        ad void
    882       1.59       dsl leavepgrp(struct proc *p)
    883        1.1       cgd {
    884       1.61       dsl 	struct pgrp *pgrp;
    885        1.1       cgd 
    886   1.94.4.5        ad 	LOCK_ASSERT(rw_write_held(&proclist_lock));
    887   1.94.4.5        ad 
    888   1.94.4.1        ad 	/*
    889   1.94.4.1        ad 	 * If there's a controlling terminal for the session, we have to
    890   1.94.4.1        ad 	 * interlock with it.  See ttread().
    891   1.94.4.1        ad 	 */
    892   1.94.4.1        ad 	mutex_enter(&proclist_mutex);
    893   1.94.4.1        ad #ifdef notyet
    894   1.94.4.1        ad 	if (p_>p_session->s_ttyvp != NULL) {
    895   1.94.4.1        ad 		tp = p->p_session->s_ttyp;
    896   1.94.4.1        ad 		mutex_enter(&tp->t_mutex);
    897   1.94.4.1        ad 	} else
    898   1.94.4.1        ad 		tp = NULL;
    899   1.94.4.1        ad #endif
    900   1.94.4.1        ad 
    901       1.61       dsl 	pgrp = p->p_pgrp;
    902       1.10   mycroft 	LIST_REMOVE(p, p_pglist);
    903       1.94        ad 	p->p_pgrp = NULL;
    904       1.61       dsl 
    905   1.94.4.1        ad #ifdef notyet
    906   1.94.4.1        ad 	if (tp != NULL)
    907   1.94.4.1        ad 		mutex_exit(&tp->t_mutex);
    908   1.94.4.1        ad #endif
    909   1.94.4.1        ad 	mutex_exit(&proclist_mutex);
    910   1.94.4.1        ad 
    911   1.94.4.1        ad 	if (LIST_EMPTY(&pgrp->pg_members))
    912   1.94.4.1        ad 		pg_delete(pgrp->pg_id);
    913       1.61       dsl }
    914       1.61       dsl 
    915   1.94.4.5        ad /*
    916   1.94.4.5        ad  * Free a process group.  Must be called with the proclist_lock write held.
    917   1.94.4.5        ad  */
    918       1.61       dsl static void
    919       1.61       dsl pg_free(pid_t pg_id)
    920       1.61       dsl {
    921       1.61       dsl 	struct pgrp *pgrp;
    922       1.61       dsl 	struct pid_table *pt;
    923       1.61       dsl 
    924   1.94.4.5        ad 	LOCK_ASSERT(rw_write_held(&proclist_lock));
    925   1.94.4.5        ad 
    926       1.61       dsl 	pt = &pid_table[pg_id & pid_tbl_mask];
    927       1.61       dsl 	pgrp = pt->pt_pgrp;
    928       1.61       dsl #ifdef DIAGNOSTIC
    929       1.63  christos 	if (__predict_false(!pgrp || pgrp->pg_id != pg_id
    930       1.63  christos 	    || !LIST_EMPTY(&pgrp->pg_members)))
    931       1.61       dsl 		panic("pg_free: process group absent or has members");
    932       1.61       dsl #endif
    933       1.61       dsl 	pt->pt_pgrp = 0;
    934       1.61       dsl 
    935       1.61       dsl 	if (!P_VALID(pt->pt_proc)) {
    936       1.61       dsl 		/* orphaned pgrp, put slot onto free list */
    937       1.61       dsl #ifdef DIAGNOSTIC
    938       1.63  christos 		if (__predict_false(P_NEXT(pt->pt_proc) & pid_tbl_mask))
    939       1.61       dsl 			panic("pg_free: process slot on free list");
    940       1.61       dsl #endif
    941   1.94.4.2        ad 		mutex_enter(&proclist_mutex);
    942       1.61       dsl 		pg_id &= pid_tbl_mask;
    943       1.61       dsl 		pt = &pid_table[last_free_pt];
    944       1.61       dsl 		pt->pt_proc = P_FREE(P_NEXT(pt->pt_proc) | pg_id);
    945   1.94.4.2        ad 		mutex_exit(&proclist_mutex);
    946       1.61       dsl 		last_free_pt = pg_id;
    947       1.61       dsl 		pid_alloc_cnt--;
    948       1.61       dsl 	}
    949       1.61       dsl 	pool_put(&pgrp_pool, pgrp);
    950        1.1       cgd }
    951        1.1       cgd 
    952        1.1       cgd /*
    953   1.94.4.5        ad  * Delete a process group.  Must be called with the proclist_lock write
    954   1.94.4.5        ad  * held.
    955        1.1       cgd  */
    956       1.61       dsl static void
    957       1.61       dsl pg_delete(pid_t pg_id)
    958       1.61       dsl {
    959       1.61       dsl 	struct pgrp *pgrp;
    960       1.61       dsl 	struct tty *ttyp;
    961       1.61       dsl 	struct session *ss;
    962   1.94.4.1        ad 	int is_pgrp_leader;
    963       1.61       dsl 
    964   1.94.4.5        ad 	LOCK_ASSERT(rw_write_held(&proclist_lock));
    965   1.94.4.5        ad 
    966       1.61       dsl 	pgrp = pid_table[pg_id & pid_tbl_mask].pt_pgrp;
    967       1.61       dsl 	if (pgrp == NULL || pgrp->pg_id != pg_id ||
    968   1.94.4.5        ad 	    !LIST_EMPTY(&pgrp->pg_members))
    969       1.61       dsl 		return;
    970       1.61       dsl 
    971       1.71        pk 	ss = pgrp->pg_session;
    972       1.71        pk 
    973       1.61       dsl 	/* Remove reference (if any) from tty to this process group */
    974       1.71        pk 	ttyp = ss->s_ttyp;
    975       1.71        pk 	if (ttyp != NULL && ttyp->t_pgrp == pgrp) {
    976       1.61       dsl 		ttyp->t_pgrp = NULL;
    977       1.71        pk #ifdef DIAGNOSTIC
    978       1.71        pk 		if (ttyp->t_session != ss)
    979       1.71        pk 			panic("pg_delete: wrong session on terminal");
    980       1.71        pk #endif
    981       1.71        pk 	}
    982       1.61       dsl 
    983       1.71        pk 	/*
    984       1.71        pk 	 * The leading process group in a session is freed
    985       1.71        pk 	 * by sessdelete() if last reference.
    986       1.71        pk 	 */
    987       1.71        pk 	is_pgrp_leader = (ss->s_sid == pgrp->pg_id);
    988       1.71        pk 	SESSRELE(ss);
    989       1.61       dsl 
    990       1.71        pk 	if (is_pgrp_leader)
    991       1.61       dsl 		return;
    992       1.61       dsl 
    993       1.61       dsl 	pg_free(pg_id);
    994       1.61       dsl }
    995       1.61       dsl 
    996       1.61       dsl /*
    997       1.61       dsl  * Delete session - called from SESSRELE when s_count becomes zero.
    998   1.94.4.5        ad  * Must be called with the proclist_lock write held.
    999       1.61       dsl  */
   1000       1.11       cgd void
   1001       1.61       dsl sessdelete(struct session *ss)
   1002        1.1       cgd {
   1003   1.94.4.5        ad 
   1004   1.94.4.5        ad 	LOCK_ASSERT(rw_write_held(&proclist_lock));
   1005   1.94.4.5        ad 
   1006       1.61       dsl 	/*
   1007       1.61       dsl 	 * We keep the pgrp with the same id as the session in
   1008       1.61       dsl 	 * order to stop a process being given the same pid.
   1009       1.61       dsl 	 * Since the pgrp holds a reference to the session, it
   1010       1.61       dsl 	 * must be a 'zombie' pgrp by now.
   1011       1.61       dsl 	 */
   1012       1.61       dsl 	pg_free(ss->s_sid);
   1013       1.77    simonb 	pool_put(&session_pool, ss);
   1014        1.1       cgd }
   1015        1.1       cgd 
   1016        1.1       cgd /*
   1017        1.1       cgd  * Adjust pgrp jobc counters when specified process changes process group.
   1018        1.1       cgd  * We count the number of processes in each process group that "qualify"
   1019        1.1       cgd  * the group for terminal job control (those with a parent in a different
   1020        1.1       cgd  * process group of the same session).  If that count reaches zero, the
   1021        1.1       cgd  * process group becomes orphaned.  Check both the specified process'
   1022        1.1       cgd  * process group and that of its children.
   1023        1.1       cgd  * entering == 0 => p is leaving specified group.
   1024        1.1       cgd  * entering == 1 => p is entering specified group.
   1025       1.68       dsl  *
   1026   1.94.4.5        ad  * Call with proclist_lock write held.
   1027        1.1       cgd  */
   1028        1.4    andrew void
   1029       1.59       dsl fixjobc(struct proc *p, struct pgrp *pgrp, int entering)
   1030        1.1       cgd {
   1031       1.39  augustss 	struct pgrp *hispgrp;
   1032       1.39  augustss 	struct session *mysession = pgrp->pg_session;
   1033       1.68       dsl 	struct proc *child;
   1034        1.1       cgd 
   1035   1.94.4.5        ad 	LOCK_ASSERT(rw_write_held(&proclist_lock));
   1036  1.94.4.12        ad 	LOCK_ASSERT(mutex_owned(&proclist_mutex));
   1037   1.94.4.5        ad 
   1038        1.1       cgd 	/*
   1039        1.1       cgd 	 * Check p's parent to see whether p qualifies its own process
   1040        1.1       cgd 	 * group; if so, adjust count for p's process group.
   1041        1.1       cgd 	 */
   1042       1.68       dsl 	hispgrp = p->p_pptr->p_pgrp;
   1043       1.68       dsl 	if (hispgrp != pgrp && hispgrp->pg_session == mysession) {
   1044   1.94.4.3        ad 		if (entering) {
   1045   1.94.4.5        ad 			mutex_enter(&p->p_smutex);
   1046   1.94.4.5        ad 			p->p_sflag &= ~PS_ORPHANPG;
   1047   1.94.4.5        ad 			mutex_exit(&p->p_smutex);
   1048        1.1       cgd 			pgrp->pg_jobc++;
   1049   1.94.4.3        ad 		} else if (--pgrp->pg_jobc == 0)
   1050        1.1       cgd 			orphanpg(pgrp);
   1051       1.26   thorpej 	}
   1052        1.1       cgd 
   1053        1.1       cgd 	/*
   1054        1.1       cgd 	 * Check this process' children to see whether they qualify
   1055        1.1       cgd 	 * their process groups; if so, adjust counts for children's
   1056        1.1       cgd 	 * process groups.
   1057        1.1       cgd 	 */
   1058       1.68       dsl 	LIST_FOREACH(child, &p->p_children, p_sibling) {
   1059       1.68       dsl 		hispgrp = child->p_pgrp;
   1060       1.68       dsl 		if (hispgrp != pgrp && hispgrp->pg_session == mysession &&
   1061       1.68       dsl 		    !P_ZOMBIE(child)) {
   1062   1.94.4.3        ad 			if (entering) {
   1063   1.94.4.5        ad 				mutex_enter(&child->p_smutex);
   1064   1.94.4.5        ad 				child->p_sflag &= ~PS_ORPHANPG;
   1065   1.94.4.5        ad 				mutex_exit(&child->p_smutex);
   1066        1.1       cgd 				hispgrp->pg_jobc++;
   1067   1.94.4.3        ad 			} else if (--hispgrp->pg_jobc == 0)
   1068        1.1       cgd 				orphanpg(hispgrp);
   1069       1.26   thorpej 		}
   1070       1.26   thorpej 	}
   1071        1.1       cgd }
   1072        1.1       cgd 
   1073       1.72  junyoung /*
   1074        1.1       cgd  * A process group has become orphaned;
   1075        1.1       cgd  * if there are any stopped processes in the group,
   1076        1.1       cgd  * hang-up all process in that group.
   1077       1.68       dsl  *
   1078   1.94.4.5        ad  * Call with proclist_lock write held.
   1079        1.1       cgd  */
   1080        1.4    andrew static void
   1081       1.59       dsl orphanpg(struct pgrp *pg)
   1082        1.1       cgd {
   1083       1.39  augustss 	struct proc *p;
   1084   1.94.4.3        ad 	int doit;
   1085   1.94.4.3        ad 
   1086   1.94.4.5        ad 	LOCK_ASSERT(rw_write_held(&proclist_lock));
   1087  1.94.4.12        ad 	LOCK_ASSERT(mutex_owned(&proclist_mutex));
   1088   1.94.4.5        ad 
   1089   1.94.4.3        ad 	doit = 0;
   1090        1.1       cgd 
   1091       1.52      matt 	LIST_FOREACH(p, &pg->pg_members, p_pglist) {
   1092   1.94.4.3        ad 		mutex_enter(&p->p_smutex);
   1093        1.1       cgd 		if (p->p_stat == SSTOP) {
   1094   1.94.4.3        ad 			doit = 1;
   1095   1.94.4.5        ad 			p->p_sflag |= PS_ORPHANPG;
   1096   1.94.4.3        ad 		}
   1097   1.94.4.3        ad 		mutex_exit(&p->p_smutex);
   1098   1.94.4.3        ad 	}
   1099   1.94.4.3        ad 
   1100   1.94.4.3        ad 	if (doit) {
   1101   1.94.4.3        ad 		LIST_FOREACH(p, &pg->pg_members, p_pglist) {
   1102   1.94.4.3        ad 			psignal(p, SIGHUP);
   1103   1.94.4.3        ad 			psignal(p, SIGCONT);
   1104        1.1       cgd 		}
   1105        1.1       cgd 	}
   1106        1.1       cgd }
   1107       1.35    bouyer 
   1108       1.61       dsl #ifdef DDB
   1109       1.61       dsl #include <ddb/db_output.h>
   1110       1.61       dsl void pidtbl_dump(void);
   1111       1.14  christos void
   1112       1.61       dsl pidtbl_dump(void)
   1113        1.1       cgd {
   1114       1.61       dsl 	struct pid_table *pt;
   1115       1.61       dsl 	struct proc *p;
   1116       1.39  augustss 	struct pgrp *pgrp;
   1117       1.61       dsl 	int id;
   1118        1.1       cgd 
   1119       1.61       dsl 	db_printf("pid table %p size %x, next %x, last %x\n",
   1120       1.61       dsl 		pid_table, pid_tbl_mask+1,
   1121       1.61       dsl 		next_free_pt, last_free_pt);
   1122       1.61       dsl 	for (pt = pid_table, id = 0; id <= pid_tbl_mask; id++, pt++) {
   1123       1.61       dsl 		p = pt->pt_proc;
   1124       1.61       dsl 		if (!P_VALID(p) && !pt->pt_pgrp)
   1125       1.61       dsl 			continue;
   1126       1.61       dsl 		db_printf("  id %x: ", id);
   1127       1.61       dsl 		if (P_VALID(p))
   1128       1.61       dsl 			db_printf("proc %p id %d (0x%x) %s\n",
   1129       1.61       dsl 				p, p->p_pid, p->p_pid, p->p_comm);
   1130       1.61       dsl 		else
   1131       1.61       dsl 			db_printf("next %x use %x\n",
   1132       1.61       dsl 				P_NEXT(p) & pid_tbl_mask,
   1133       1.61       dsl 				P_NEXT(p) & ~pid_tbl_mask);
   1134       1.61       dsl 		if ((pgrp = pt->pt_pgrp)) {
   1135       1.61       dsl 			db_printf("\tsession %p, sid %d, count %d, login %s\n",
   1136       1.61       dsl 			    pgrp->pg_session, pgrp->pg_session->s_sid,
   1137       1.61       dsl 			    pgrp->pg_session->s_count,
   1138       1.61       dsl 			    pgrp->pg_session->s_login);
   1139       1.61       dsl 			db_printf("\tpgrp %p, pg_id %d, pg_jobc %d, members %p\n",
   1140       1.61       dsl 			    pgrp, pgrp->pg_id, pgrp->pg_jobc,
   1141       1.61       dsl 			    pgrp->pg_members.lh_first);
   1142       1.61       dsl 			for (p = pgrp->pg_members.lh_first; p != 0;
   1143       1.61       dsl 			    p = p->p_pglist.le_next) {
   1144       1.72  junyoung 				db_printf("\t\tpid %d addr %p pgrp %p %s\n",
   1145       1.61       dsl 				    p->p_pid, p, p->p_pgrp, p->p_comm);
   1146       1.10   mycroft 			}
   1147        1.1       cgd 		}
   1148        1.1       cgd 	}
   1149        1.1       cgd }
   1150       1.61       dsl #endif /* DDB */
   1151       1.48      yamt 
   1152       1.48      yamt #ifdef KSTACK_CHECK_MAGIC
   1153       1.48      yamt #include <sys/user.h>
   1154       1.48      yamt 
   1155       1.48      yamt #define	KSTACK_MAGIC	0xdeadbeaf
   1156       1.48      yamt 
   1157       1.48      yamt /* XXX should be per process basis? */
   1158       1.48      yamt int kstackleftmin = KSTACK_SIZE;
   1159       1.50     enami int kstackleftthres = KSTACK_SIZE / 8; /* warn if remaining stack is
   1160       1.50     enami 					  less than this */
   1161       1.48      yamt 
   1162       1.48      yamt void
   1163       1.56      yamt kstack_setup_magic(const struct lwp *l)
   1164       1.48      yamt {
   1165       1.85     perry 	uint32_t *ip;
   1166       1.85     perry 	uint32_t const *end;
   1167       1.48      yamt 
   1168       1.56      yamt 	KASSERT(l != NULL);
   1169       1.56      yamt 	KASSERT(l != &lwp0);
   1170       1.48      yamt 
   1171       1.48      yamt 	/*
   1172       1.48      yamt 	 * fill all the stack with magic number
   1173       1.48      yamt 	 * so that later modification on it can be detected.
   1174       1.48      yamt 	 */
   1175       1.85     perry 	ip = (uint32_t *)KSTACK_LOWEST_ADDR(l);
   1176       1.85     perry 	end = (uint32_t *)((caddr_t)KSTACK_LOWEST_ADDR(l) + KSTACK_SIZE);
   1177       1.48      yamt 	for (; ip < end; ip++) {
   1178       1.48      yamt 		*ip = KSTACK_MAGIC;
   1179       1.48      yamt 	}
   1180       1.48      yamt }
   1181       1.48      yamt 
   1182       1.48      yamt void
   1183       1.56      yamt kstack_check_magic(const struct lwp *l)
   1184       1.48      yamt {
   1185       1.85     perry 	uint32_t const *ip, *end;
   1186       1.48      yamt 	int stackleft;
   1187       1.48      yamt 
   1188       1.56      yamt 	KASSERT(l != NULL);
   1189       1.48      yamt 
   1190       1.48      yamt 	/* don't check proc0 */ /*XXX*/
   1191       1.56      yamt 	if (l == &lwp0)
   1192       1.48      yamt 		return;
   1193       1.48      yamt 
   1194       1.48      yamt #ifdef __MACHINE_STACK_GROWS_UP
   1195       1.48      yamt 	/* stack grows upwards (eg. hppa) */
   1196       1.85     perry 	ip = (uint32_t *)((caddr_t)KSTACK_LOWEST_ADDR(l) + KSTACK_SIZE);
   1197       1.85     perry 	end = (uint32_t *)KSTACK_LOWEST_ADDR(l);
   1198       1.48      yamt 	for (ip--; ip >= end; ip--)
   1199       1.48      yamt 		if (*ip != KSTACK_MAGIC)
   1200       1.48      yamt 			break;
   1201       1.72  junyoung 
   1202       1.56      yamt 	stackleft = (caddr_t)KSTACK_LOWEST_ADDR(l) + KSTACK_SIZE - (caddr_t)ip;
   1203       1.48      yamt #else /* __MACHINE_STACK_GROWS_UP */
   1204       1.48      yamt 	/* stack grows downwards (eg. i386) */
   1205       1.85     perry 	ip = (uint32_t *)KSTACK_LOWEST_ADDR(l);
   1206       1.85     perry 	end = (uint32_t *)((caddr_t)KSTACK_LOWEST_ADDR(l) + KSTACK_SIZE);
   1207       1.48      yamt 	for (; ip < end; ip++)
   1208       1.48      yamt 		if (*ip != KSTACK_MAGIC)
   1209       1.48      yamt 			break;
   1210       1.48      yamt 
   1211       1.93  christos 	stackleft = ((const char *)ip) - (const char *)KSTACK_LOWEST_ADDR(l);
   1212       1.48      yamt #endif /* __MACHINE_STACK_GROWS_UP */
   1213       1.48      yamt 
   1214       1.48      yamt 	if (kstackleftmin > stackleft) {
   1215       1.48      yamt 		kstackleftmin = stackleft;
   1216       1.48      yamt 		if (stackleft < kstackleftthres)
   1217       1.56      yamt 			printf("warning: kernel stack left %d bytes"
   1218       1.56      yamt 			    "(pid %u:lid %u)\n", stackleft,
   1219       1.56      yamt 			    (u_int)l->l_proc->p_pid, (u_int)l->l_lid);
   1220       1.48      yamt 	}
   1221       1.48      yamt 
   1222       1.48      yamt 	if (stackleft <= 0) {
   1223       1.56      yamt 		panic("magic on the top of kernel stack changed for "
   1224       1.56      yamt 		    "pid %u, lid %u: maybe kernel stack overflow",
   1225       1.56      yamt 		    (u_int)l->l_proc->p_pid, (u_int)l->l_lid);
   1226       1.48      yamt 	}
   1227       1.48      yamt }
   1228       1.50     enami #endif /* KSTACK_CHECK_MAGIC */
   1229       1.79      yamt 
   1230   1.94.4.1        ad /*
   1231   1.94.4.1        ad  * XXXSMP this is bust, it grabs a read lock and then messes about
   1232   1.94.4.1        ad  * with allproc.
   1233   1.94.4.1        ad  */
   1234       1.79      yamt int
   1235       1.79      yamt proclist_foreach_call(struct proclist *list,
   1236       1.79      yamt     int (*callback)(struct proc *, void *arg), void *arg)
   1237       1.79      yamt {
   1238       1.79      yamt 	struct proc marker;
   1239       1.79      yamt 	struct proc *p;
   1240       1.79      yamt 	struct lwp * const l = curlwp;
   1241       1.79      yamt 	int ret = 0;
   1242       1.79      yamt 
   1243       1.79      yamt 	marker.p_flag = P_MARKER;
   1244       1.79      yamt 	PHOLD(l);
   1245   1.94.4.1        ad 	rw_enter(&proclist_lock, RW_READER);
   1246       1.79      yamt 	for (p = LIST_FIRST(list); ret == 0 && p != NULL;) {
   1247       1.79      yamt 		if (p->p_flag & P_MARKER) {
   1248       1.79      yamt 			p = LIST_NEXT(p, p_list);
   1249       1.79      yamt 			continue;
   1250       1.79      yamt 		}
   1251       1.79      yamt 		LIST_INSERT_AFTER(p, &marker, p_list);
   1252       1.79      yamt 		ret = (*callback)(p, arg);
   1253   1.94.4.1        ad 		KASSERT(rw_read_held(&proclist_lock));
   1254       1.79      yamt 		p = LIST_NEXT(&marker, p_list);
   1255       1.79      yamt 		LIST_REMOVE(&marker, p_list);
   1256       1.79      yamt 	}
   1257   1.94.4.1        ad 	rw_exit(&proclist_lock);
   1258       1.79      yamt 	PRELE(l);
   1259       1.79      yamt 
   1260       1.79      yamt 	return ret;
   1261       1.79      yamt }
   1262       1.86      yamt 
   1263       1.86      yamt int
   1264       1.86      yamt proc_vmspace_getref(struct proc *p, struct vmspace **vm)
   1265       1.86      yamt {
   1266       1.86      yamt 
   1267       1.86      yamt 	/* XXXCDC: how should locking work here? */
   1268       1.86      yamt 
   1269       1.87      yamt 	/* curproc exception is for coredump. */
   1270       1.87      yamt 
   1271   1.94.4.5        ad 	if ((p != curproc && (p->p_sflag & PS_WEXIT) != 0) ||
   1272       1.86      yamt 	    (p->p_vmspace->vm_refcnt < 1)) { /* XXX */
   1273       1.86      yamt 		return EFAULT;
   1274       1.86      yamt 	}
   1275       1.86      yamt 
   1276       1.86      yamt 	uvmspace_addref(p->p_vmspace);
   1277       1.86      yamt 	*vm = p->p_vmspace;
   1278       1.86      yamt 
   1279       1.86      yamt 	return 0;
   1280       1.86      yamt }
   1281       1.94        ad 
   1282       1.94        ad /*
   1283       1.94        ad  * Acquire a write lock on the process credential.
   1284       1.94        ad  */
   1285       1.94        ad void
   1286   1.94.4.5        ad proc_crmod_enter(void)
   1287       1.94        ad {
   1288   1.94.4.5        ad 	struct lwp *l = curlwp;
   1289   1.94.4.5        ad 	struct proc *p = l->l_proc;
   1290  1.94.4.14        ad 	struct plimit *lim;
   1291   1.94.4.5        ad 	kauth_cred_t oc;
   1292  1.94.4.14        ad 	char *cn;
   1293   1.94.4.5        ad 
   1294   1.94.4.5        ad 	mutex_enter(&p->p_mutex);
   1295   1.94.4.5        ad 
   1296   1.94.4.5        ad 	/* Ensure the LWP cached credentials are up to date. */
   1297   1.94.4.5        ad 	if ((oc = l->l_cred) != p->p_cred) {
   1298   1.94.4.5        ad 		kauth_cred_hold(p->p_cred);
   1299   1.94.4.5        ad 		l->l_cred = p->p_cred;
   1300   1.94.4.5        ad 		kauth_cred_free(oc);
   1301   1.94.4.5        ad 	}
   1302  1.94.4.14        ad 
   1303  1.94.4.14        ad 	/* Reset what needs to be reset in plimit. */
   1304  1.94.4.14        ad 	lim = p->p_limit;
   1305  1.94.4.14        ad 	if (lim->pl_corename != defcorename) {
   1306  1.94.4.14        ad 		if (lim->p_refcnt > 1 &&
   1307  1.94.4.14        ad 		    (lim->p_lflags & PL_SHAREMOD) == 0) {
   1308  1.94.4.14        ad 			p->p_limit = limcopy(p);
   1309  1.94.4.14        ad 			limfree(lim);
   1310  1.94.4.14        ad 			lim = p->p_limit;
   1311  1.94.4.14        ad 		}
   1312  1.94.4.14        ad 		simple_lock(&lim->p_slock);
   1313  1.94.4.14        ad 		cn = lim->pl_corename;
   1314  1.94.4.14        ad 		lim->pl_corename = defcorename;
   1315  1.94.4.14        ad 		simple_unlock(&lim->p_slock);
   1316  1.94.4.14        ad 		if (cn != defcorename)
   1317  1.94.4.14        ad 			free(cn, M_TEMP);
   1318  1.94.4.14        ad 	}
   1319       1.94        ad }
   1320       1.94        ad 
   1321       1.94        ad /*
   1322   1.94.4.3        ad  * Set in a new process credential, and drop the write lock.  The credential
   1323   1.94.4.3        ad  * must have a reference already.  Optionally, free a no-longer required
   1324   1.94.4.3        ad  * credential.  The scheduler also needs to inspect p_cred, so we also
   1325   1.94.4.3        ad  * briefly acquire the sched state mutex.
   1326       1.94        ad  */
   1327       1.94        ad void
   1328  1.94.4.14        ad proc_crmod_leave(kauth_cred_t scred, kauth_cred_t fcred, boolean_t sugid)
   1329       1.94        ad {
   1330   1.94.4.5        ad 	struct lwp *l = curlwp;
   1331   1.94.4.5        ad 	struct proc *p = l->l_proc;
   1332   1.94.4.5        ad 	kauth_cred_t oc;
   1333       1.94        ad 
   1334  1.94.4.14        ad 	/* Is there a new credential to set in? */
   1335  1.94.4.14        ad 	if (scred != NULL) {
   1336  1.94.4.14        ad 		mutex_enter(&p->p_smutex);
   1337  1.94.4.14        ad 		p->p_cred = scred;
   1338  1.94.4.14        ad 		mutex_exit(&p->p_smutex);
   1339   1.94.4.5        ad 
   1340  1.94.4.14        ad 		/* Ensure the LWP cached credentials are up to date. */
   1341  1.94.4.14        ad 		if ((oc = l->l_cred) != scred) {
   1342  1.94.4.14        ad 			kauth_cred_hold(scred);
   1343  1.94.4.14        ad 			l->l_cred = scred;
   1344  1.94.4.14        ad 		}
   1345  1.94.4.14        ad 	} else
   1346  1.94.4.14        ad 		oc = NULL;	/* XXXgcc */
   1347  1.94.4.14        ad 
   1348  1.94.4.14        ad 	if (sugid) {
   1349  1.94.4.14        ad 		/*
   1350  1.94.4.14        ad 		 * Mark process as having changed credentials, stops
   1351  1.94.4.14        ad 		 * tracing etc.
   1352  1.94.4.14        ad 		 */
   1353  1.94.4.14        ad 		p->p_flag |= P_SUGID;
   1354   1.94.4.5        ad 	}
   1355   1.94.4.5        ad 
   1356   1.94.4.5        ad 	mutex_exit(&p->p_mutex);
   1357  1.94.4.14        ad 
   1358  1.94.4.14        ad 	/* If there is a credential to be released, free it now. */
   1359  1.94.4.14        ad 	if (fcred != NULL) {
   1360  1.94.4.14        ad 		KASSERT(scred != NULL);
   1361  1.94.4.14        ad 		kauth_cred_free(fcred);
   1362  1.94.4.14        ad 		if (oc != scred)
   1363  1.94.4.14        ad 			kauth_cred_free(oc);
   1364  1.94.4.14        ad 	}
   1365   1.94.4.5        ad }
   1366   1.94.4.5        ad 
   1367   1.94.4.5        ad /*
   1368   1.94.4.5        ad  * Acquire a reference on a process, to prevent it from exiting or execing.
   1369   1.94.4.5        ad  */
   1370   1.94.4.5        ad int
   1371   1.94.4.5        ad proc_addref(struct proc *p)
   1372   1.94.4.5        ad {
   1373   1.94.4.5        ad 
   1374   1.94.4.5        ad 	LOCK_ASSERT(mutex_owned(&p->p_mutex));
   1375   1.94.4.5        ad 
   1376   1.94.4.5        ad 	if (p->p_refcnt <= 0)
   1377   1.94.4.5        ad 		return EAGAIN;
   1378   1.94.4.5        ad 	p->p_refcnt++;
   1379   1.94.4.5        ad 
   1380   1.94.4.5        ad 	return 0;
   1381   1.94.4.5        ad }
   1382   1.94.4.5        ad 
   1383   1.94.4.5        ad /*
   1384   1.94.4.5        ad  * Release a reference on a process.
   1385   1.94.4.5        ad  */
   1386   1.94.4.5        ad void
   1387   1.94.4.5        ad proc_delref(struct proc *p)
   1388   1.94.4.5        ad {
   1389   1.94.4.5        ad 
   1390   1.94.4.5        ad 	LOCK_ASSERT(mutex_owned(&p->p_mutex));
   1391   1.94.4.5        ad 
   1392   1.94.4.5        ad 	if (p->p_refcnt < 0) {
   1393   1.94.4.5        ad 		if (++p->p_refcnt == 0)
   1394   1.94.4.7        ad 			cv_broadcast(&p->p_refcv);
   1395   1.94.4.5        ad 	} else {
   1396   1.94.4.5        ad 		p->p_refcnt--;
   1397   1.94.4.5        ad 		KASSERT(p->p_refcnt != 0);
   1398   1.94.4.5        ad 	}
   1399   1.94.4.5        ad }
   1400   1.94.4.5        ad 
   1401   1.94.4.5        ad /*
   1402   1.94.4.5        ad  * Wait for all references on the process to drain, and prevent new
   1403   1.94.4.5        ad  * references from being acquired.
   1404   1.94.4.5        ad  */
   1405   1.94.4.5        ad void
   1406   1.94.4.5        ad proc_drainrefs(struct proc *p)
   1407   1.94.4.5        ad {
   1408   1.94.4.5        ad 
   1409   1.94.4.5        ad 	LOCK_ASSERT(mutex_owned(&p->p_mutex));
   1410   1.94.4.5        ad 	KASSERT(p->p_refcnt > 0);
   1411   1.94.4.5        ad 
   1412   1.94.4.5        ad 	/*
   1413   1.94.4.5        ad 	 * The process itself holds the last reference.  Once it's released,
   1414   1.94.4.5        ad 	 * no new references will be granted.  If we have already locked out
   1415   1.94.4.5        ad 	 * new references (refcnt <= 0), potentially due to a failed exec,
   1416   1.94.4.5        ad 	 * there is nothing more to do.
   1417   1.94.4.5        ad 	 */
   1418   1.94.4.5        ad 	p->p_refcnt = 1 - p->p_refcnt;
   1419   1.94.4.5        ad 	while (p->p_refcnt != 0)
   1420   1.94.4.5        ad 		cv_wait(&p->p_refcv, &p->p_mutex);
   1421       1.94        ad }
   1422   1.94.4.6        ad 
   1423   1.94.4.6        ad /*
   1424   1.94.4.6        ad  * proc_specific_key_create --
   1425   1.94.4.6        ad  *	Create a key for subsystem proc-specific data.
   1426   1.94.4.6        ad  */
   1427   1.94.4.6        ad int
   1428   1.94.4.6        ad proc_specific_key_create(specificdata_key_t *keyp, specificdata_dtor_t dtor)
   1429   1.94.4.6        ad {
   1430   1.94.4.6        ad 
   1431   1.94.4.6        ad 	return (specificdata_key_create(proc_specificdata_domain, keyp, dtor));
   1432   1.94.4.6        ad }
   1433   1.94.4.6        ad 
   1434   1.94.4.6        ad /*
   1435   1.94.4.6        ad  * proc_specific_key_delete --
   1436   1.94.4.6        ad  *	Delete a key for subsystem proc-specific data.
   1437   1.94.4.6        ad  */
   1438   1.94.4.6        ad void
   1439   1.94.4.6        ad proc_specific_key_delete(specificdata_key_t key)
   1440   1.94.4.6        ad {
   1441   1.94.4.6        ad 
   1442   1.94.4.6        ad 	specificdata_key_delete(proc_specificdata_domain, key);
   1443   1.94.4.6        ad }
   1444   1.94.4.6        ad 
   1445   1.94.4.6        ad /*
   1446   1.94.4.6        ad  * proc_initspecific --
   1447   1.94.4.6        ad  *	Initialize a proc's specificdata container.
   1448   1.94.4.6        ad  */
   1449   1.94.4.6        ad void
   1450   1.94.4.6        ad proc_initspecific(struct proc *p)
   1451   1.94.4.6        ad {
   1452   1.94.4.6        ad 	int error;
   1453   1.94.4.6        ad 
   1454   1.94.4.6        ad 	error = specificdata_init(proc_specificdata_domain, &p->p_specdataref);
   1455   1.94.4.6        ad 	KASSERT(error == 0);
   1456   1.94.4.6        ad }
   1457   1.94.4.6        ad 
   1458   1.94.4.6        ad /*
   1459   1.94.4.6        ad  * proc_finispecific --
   1460   1.94.4.6        ad  *	Finalize a proc's specificdata container.
   1461   1.94.4.6        ad  */
   1462   1.94.4.6        ad void
   1463   1.94.4.6        ad proc_finispecific(struct proc *p)
   1464   1.94.4.6        ad {
   1465   1.94.4.6        ad 
   1466   1.94.4.6        ad 	specificdata_fini(proc_specificdata_domain, &p->p_specdataref);
   1467   1.94.4.6        ad }
   1468   1.94.4.6        ad 
   1469   1.94.4.6        ad /*
   1470   1.94.4.6        ad  * proc_getspecific --
   1471   1.94.4.6        ad  *	Return proc-specific data corresponding to the specified key.
   1472   1.94.4.6        ad  */
   1473   1.94.4.6        ad void *
   1474   1.94.4.6        ad proc_getspecific(struct proc *p, specificdata_key_t key)
   1475   1.94.4.6        ad {
   1476   1.94.4.6        ad 
   1477   1.94.4.6        ad 	return (specificdata_getspecific(proc_specificdata_domain,
   1478   1.94.4.6        ad 					 &p->p_specdataref, key));
   1479   1.94.4.6        ad }
   1480   1.94.4.6        ad 
   1481   1.94.4.6        ad /*
   1482   1.94.4.6        ad  * proc_setspecific --
   1483   1.94.4.6        ad  *	Set proc-specific data corresponding to the specified key.
   1484   1.94.4.6        ad  */
   1485   1.94.4.6        ad void
   1486   1.94.4.6        ad proc_setspecific(struct proc *p, specificdata_key_t key, void *data)
   1487   1.94.4.6        ad {
   1488   1.94.4.6        ad 
   1489   1.94.4.6        ad 	specificdata_setspecific(proc_specificdata_domain,
   1490   1.94.4.6        ad 				 &p->p_specdataref, key, data);
   1491   1.94.4.6        ad }
   1492