kernel - Make kern_proc cache-friendly

[dragonfly.git] / sys / kern / kern_proc.c

/*
 * Copyright (c) 1982, 1986, 1989, 1991, 1993
 *      The Regents of the University of California.  All rights reserved.
 *
 * 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. 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.
 */

#include <sys/param.h>
#include <sys/systm.h>
#include <sys/kernel.h>
#include <sys/sysctl.h>
#include <sys/malloc.h>
#include <sys/proc.h>
#include <sys/vnode.h>
#include <sys/jail.h>
#include <sys/filedesc.h>
#include <sys/tty.h>
#include <sys/dsched.h>
#include <sys/signalvar.h>
#include <sys/spinlock.h>
#include <sys/random.h>
#include <sys/vnode.h>
#include <vm/vm.h>
#include <sys/lock.h>
#include <vm/pmap.h>
#include <vm/vm_map.h>
#include <sys/user.h>
#include <machine/smp.h>

#include <sys/refcount.h>
#include <sys/spinlock2.h>
#include <sys/mplock2.h>

/*
 * Hash table size must be a power of two and is not currently dynamically
 * sized.  There is a trade-off between the linear scans which must iterate
 * all HSIZE elements and the number of elements which might accumulate
 * within each hash chain.
 */
#define ALLPROC_HSIZE   256
#define ALLPROC_HMASK   (ALLPROC_HSIZE - 1)
#define ALLPROC_HASH(pid)       (pid & ALLPROC_HMASK)
#define PGRP_HASH(pid)  (pid & ALLPROC_HMASK)
#define SESS_HASH(pid)  (pid & ALLPROC_HMASK)

/*
 * pid_doms[] management, used to control how quickly a PID can be recycled.
 * Must be a multiple of ALLPROC_HSIZE for the proc_makepid() inner loops.
 *
 * WARNING! PIDDOM_DELAY should not be defined > 20 or so unless you change
 *          the array from int8_t's to int16_t's.
 */
#define PIDDOM_COUNT    10      /* 10 pids per domain - reduce array size */
#define PIDDOM_DELAY    10      /* min 10 seconds after exit before reuse */
#define PIDSEL_DOMAINS  (PID_MAX / PIDDOM_COUNT / ALLPROC_HSIZE * ALLPROC_HSIZE)

/* Used by libkvm */
int allproc_hsize = ALLPROC_HSIZE;

LIST_HEAD(pidhashhead, proc);

static MALLOC_DEFINE(M_PGRP, "pgrp", "process group header");
MALLOC_DEFINE(M_SESSION, "session", "session header");
MALLOC_DEFINE(M_PROC, "proc", "Proc structures");
MALLOC_DEFINE(M_LWP, "lwp", "lwp structures");
MALLOC_DEFINE(M_SUBPROC, "subproc", "Proc sub-structures");

int ps_showallprocs = 1;
static int ps_showallthreads = 1;
SYSCTL_INT(_security, OID_AUTO, ps_showallprocs, CTLFLAG_RW,
    &ps_showallprocs, 0,
    "Unprivileged processes can see processes with different UID/GID");
SYSCTL_INT(_security, OID_AUTO, ps_showallthreads, CTLFLAG_RW,
    &ps_showallthreads, 0,
    "Unprivileged processes can see kernel threads");
static u_int pid_domain_skips;
SYSCTL_UINT(_kern, OID_AUTO, pid_domain_skips, CTLFLAG_RW,
    &pid_domain_skips, 0,
    "Number of pid_doms[] skipped");
static u_int pid_inner_skips;
SYSCTL_UINT(_kern, OID_AUTO, pid_inner_skips, CTLFLAG_RW,
    &pid_inner_skips, 0,
    "Number of pid_doms[] skipped");

static void orphanpg(struct pgrp *pg);
static void proc_makepid(struct proc *p, int random_offset);

/*
 * Process related lists (for proc_token, allproc, allpgrp, and allsess)
 */
typedef struct procglob procglob_t;

static procglob_t       procglob[ALLPROC_HSIZE];

/*
 * We try our best to avoid recycling a PID too quickly.  We do this by
 * storing (uint8_t)time_second in the related pid domain on-reap and then
 * using that to skip-over the domain on-allocate.
 *
 * This array has to be fairly large to support a high fork/exec rate.
 * We want ~100,000 entries or so to support a 10-second reuse latency
 * at 10,000 execs/second, worst case.  Best-case multiply by PIDDOM_COUNT
 * (approximately 100,000 execs/second).
 */
static uint8_t pid_doms[PIDSEL_DOMAINS];        /* ~100,000 entries */

/*
 * Random component to nextpid generation.  We mix in a random factor to make
 * it a little harder to predict.  We sanity check the modulus value to avoid
 * doing it in critical paths.  Don't let it be too small or we pointlessly
 * waste randomness entropy, and don't let it be impossibly large.  Using a
 * modulus that is too big causes a LOT more process table scans and slows
 * down fork processing as the pidchecked caching is defeated.
 */
static int randompid = 0;

/*
 * No requirements.
 */
static int
sysctl_kern_randompid(SYSCTL_HANDLER_ARGS)
{
        int error, pid;

        pid = randompid;
        error = sysctl_handle_int(oidp, &pid, 0, req);
        if (error || !req->newptr)
                return (error);
        if (pid < 0 || pid > PID_MAX - 100)     /* out of range */
                pid = PID_MAX - 100;
        else if (pid < 2)                       /* NOP */
                pid = 0;
        else if (pid < 100)                     /* Make it reasonable */
                pid = 100;
        randompid = pid;
        return (error);
}

SYSCTL_PROC(_kern, OID_AUTO, randompid, CTLTYPE_INT|CTLFLAG_RW,
            0, 0, sysctl_kern_randompid, "I", "Random PID modulus");

/*
 * Initialize global process hashing structures.
 *
 * These functions are ONLY called from the low level boot code and do
 * not lock their operations.
 */
void
procinit(void)
{
        u_long i;

        /*
         * Avoid unnecessary stalls due to pid_doms[] values all being
         * the same.  Make sure that the allocation of pid 1 and pid 2
         * succeeds.
         */
        for (i = 0; i < PIDSEL_DOMAINS; ++i)
                pid_doms[i] = (int8_t)i - (int8_t)(PIDDOM_DELAY + 1);

        /*
         * Other misc init.
         */
        for (i = 0; i < ALLPROC_HSIZE; ++i) {
                procglob_t *prg = &procglob[i];
                LIST_INIT(&prg->allproc);
                LIST_INIT(&prg->allsess);
                LIST_INIT(&prg->allpgrp);
                lwkt_token_init(&prg->proc_token, "allproc");
        }
        uihashinit();
}

void
procinsertinit(struct proc *p)
{
        LIST_INSERT_HEAD(&procglob[ALLPROC_HASH(p->p_pid)].allproc,
                         p, p_list);
}

void
pgrpinsertinit(struct pgrp *pg)
{
        LIST_INSERT_HEAD(&procglob[ALLPROC_HASH(pg->pg_id)].allpgrp,
                         pg, pg_list);
}

void
sessinsertinit(struct session *sess)
{
        LIST_INSERT_HEAD(&procglob[ALLPROC_HASH(sess->s_sid)].allsess,
                         sess, s_list);
}

/*
 * Process hold/release support functions.  Called via the PHOLD(),
 * PRELE(), and PSTALL() macros.
 *
 * p->p_lock is a simple hold count with a waiting interlock.  No wakeup()
 * is issued unless someone is actually waiting for the process.
 *
 * Most holds are short-term, allowing a process scan or other similar
 * operation to access a proc structure without it getting ripped out from
 * under us.  procfs and process-list sysctl ops also use the hold function
 * interlocked with various p_flags to keep the vmspace intact when reading
 * or writing a user process's address space.
 *
 * There are two situations where a hold count can be longer.  Exiting lwps
 * hold the process until the lwp is reaped, and the parent will hold the
 * child during vfork()/exec() sequences while the child is marked P_PPWAIT.
 *
 * The kernel waits for the hold count to drop to 0 (or 1 in some cases) at
 * various critical points in the fork/exec and exit paths before proceeding.
 */
#define PLOCK_ZOMB      0x20000000
#define PLOCK_WAITING   0x40000000
#define PLOCK_MASK      0x1FFFFFFF

void
pstall(struct proc *p, const char *wmesg, int count)
{
        int o;
        int n;

        for (;;) {
                o = p->p_lock;
                cpu_ccfence();
                if ((o & PLOCK_MASK) <= count)
                        break;
                n = o | PLOCK_WAITING;
                tsleep_interlock(&p->p_lock, 0);

                /*
                 * If someone is trying to single-step the process during
                 * an exec or an exit they can deadlock us because procfs
                 * sleeps with the process held.
                 */
                if (p->p_stops) {
                        if (p->p_flags & P_INEXEC) {
                                wakeup(&p->p_stype);
                        } else if (p->p_flags & P_POSTEXIT) {
                                spin_lock(&p->p_spin);
                                p->p_stops = 0;
                                p->p_step = 0;
                                spin_unlock(&p->p_spin);
                                wakeup(&p->p_stype);
                        }
                }

                if (atomic_cmpset_int(&p->p_lock, o, n)) {
                        tsleep(&p->p_lock, PINTERLOCKED, wmesg, 0);
                }
        }
}

void
phold(struct proc *p)
{
        atomic_add_int(&p->p_lock, 1);
}

/*
 * WARNING!  On last release (p) can become instantly invalid due to
 *           MP races.
 */
void
prele(struct proc *p)
{
        int o;
        int n;

        /*
         * Fast path
         */
        if (atomic_cmpset_int(&p->p_lock, 1, 0))
                return;

        /*
         * Slow path
         */
        for (;;) {
                o = p->p_lock;
                KKASSERT((o & PLOCK_MASK) > 0);
                cpu_ccfence();
                n = (o - 1) & ~PLOCK_WAITING;
                if (atomic_cmpset_int(&p->p_lock, o, n)) {
                        if (o & PLOCK_WAITING)
                                wakeup(&p->p_lock);
                        break;
                }
        }
}

/*
 * Hold and flag serialized for zombie reaping purposes.
 *
 * This function will fail if it has to block, returning non-zero with
 * neither the flag set or the hold count bumped.  Note that we must block
 * without holding a ref, meaning that the caller must ensure that (p)
 * remains valid through some other interlock (typically on its parent
 * process's p_token).
 *
 * Zero is returned on success.  The hold count will be incremented and
 * the serialization flag acquired.  Note that serialization is only against
 * other pholdzomb() calls, not against phold() calls.
 */
int
pholdzomb(struct proc *p)
{
        int o;
        int n;

        /*
         * Fast path
         */
        if (atomic_cmpset_int(&p->p_lock, 0, PLOCK_ZOMB | 1))
                return(0);

        /*
         * Slow path
         */
        for (;;) {
                o = p->p_lock;
                cpu_ccfence();
                if ((o & PLOCK_ZOMB) == 0) {
                        n = (o + 1) | PLOCK_ZOMB;
                        if (atomic_cmpset_int(&p->p_lock, o, n))
                                return(0);
                } else {
                        KKASSERT((o & PLOCK_MASK) > 0);
                        n = o | PLOCK_WAITING;
                        tsleep_interlock(&p->p_lock, 0);
                        if (atomic_cmpset_int(&p->p_lock, o, n)) {
                                tsleep(&p->p_lock, PINTERLOCKED, "phldz", 0);
                                /* (p) can be ripped out at this point */
                                return(1);
                        }
                }
        }
}

/*
 * Release PLOCK_ZOMB and the hold count, waking up any waiters.
 *
 * WARNING!  On last release (p) can become instantly invalid due to
 *           MP races.
 */
void
prelezomb(struct proc *p)
{
        int o;
        int n;

        /*
         * Fast path
         */
        if (atomic_cmpset_int(&p->p_lock, PLOCK_ZOMB | 1, 0))
                return;

        /*
         * Slow path
         */
        KKASSERT(p->p_lock & PLOCK_ZOMB);
        for (;;) {
                o = p->p_lock;
                KKASSERT((o & PLOCK_MASK) > 0);
                cpu_ccfence();
                n = (o - 1) & ~(PLOCK_ZOMB | PLOCK_WAITING);
                if (atomic_cmpset_int(&p->p_lock, o, n)) {
                        if (o & PLOCK_WAITING)
                                wakeup(&p->p_lock);
                        break;
                }
        }
}

/*
 * Is p an inferior of the current process?
 *
 * No requirements.
 */
int
inferior(struct proc *p)
{
        struct proc *p2;

        PHOLD(p);
        lwkt_gettoken_shared(&p->p_token);
        while (p != curproc) {
                if (p->p_pid == 0) {
                        lwkt_reltoken(&p->p_token);
                        return (0);
                }
                p2 = p->p_pptr;
                PHOLD(p2);
                lwkt_reltoken(&p->p_token);
                PRELE(p);
                lwkt_gettoken_shared(&p2->p_token);
                p = p2;
        }
        lwkt_reltoken(&p->p_token);
        PRELE(p);

        return (1);
}

/*
 * Locate a process by number.  The returned process will be referenced and
 * must be released with PRELE().
 *
 * No requirements.
 */
struct proc *
pfind(pid_t pid)
{
        struct proc *p = curproc;
        procglob_t *prg;
        int n;

        /*
         * Shortcut the current process
         */
        if (p && p->p_pid == pid) {
                PHOLD(p);
                return (p);
        }

        /*
         * Otherwise find it in the hash table.
         */
        n = ALLPROC_HASH(pid);
        prg = &procglob[n];

        lwkt_gettoken_shared(&prg->proc_token);
        LIST_FOREACH(p, &prg->allproc, p_list) {
                if (p->p_stat == SZOMB)
                        continue;
                if (p->p_pid == pid) {
                        PHOLD(p);
                        lwkt_reltoken(&prg->proc_token);
                        return (p);
                }
        }
        lwkt_reltoken(&prg->proc_token);

        return (NULL);
}

/*
 * Locate a process by number.  The returned process is NOT referenced.
 * The result will not be stable and is typically only used to validate
 * against a process that the caller has in-hand.
 *
 * No requirements.
 */
struct proc *
pfindn(pid_t pid)
{
        struct proc *p = curproc;
        procglob_t *prg;
        int n;

        /*
         * Shortcut the current process
         */
        if (p && p->p_pid == pid)
                return (p);

        /*
         * Otherwise find it in the hash table.
         */
        n = ALLPROC_HASH(pid);
        prg = &procglob[n];

        lwkt_gettoken_shared(&prg->proc_token);
        LIST_FOREACH(p, &prg->allproc, p_list) {
                if (p->p_stat == SZOMB)
                        continue;
                if (p->p_pid == pid) {
                        lwkt_reltoken(&prg->proc_token);
                        return (p);
                }
        }
        lwkt_reltoken(&prg->proc_token);

        return (NULL);
}

/*
 * Locate a process on the zombie list.  Return a process or NULL.
 * The returned process will be referenced and the caller must release
 * it with PRELE().
 *
 * No other requirements.
 */
struct proc *
zpfind(pid_t pid)
{
        struct proc *p = curproc;
        procglob_t *prg;
        int n;

        /*
         * Shortcut the current process
         */
        if (p && p->p_pid == pid) {
                PHOLD(p);
                return (p);
        }

        /*
         * Otherwise find it in the hash table.
         */
        n = ALLPROC_HASH(pid);
        prg = &procglob[n];

        lwkt_gettoken_shared(&prg->proc_token);
        LIST_FOREACH(p, &prg->allproc, p_list) {
                if (p->p_stat != SZOMB)
                        continue;
                if (p->p_pid == pid) {
                        PHOLD(p);
                        lwkt_reltoken(&prg->proc_token);
                        return (p);
                }
        }
        lwkt_reltoken(&prg->proc_token);

        return (NULL);
}


void
pgref(struct pgrp *pgrp)
{
        refcount_acquire(&pgrp->pg_refs);
}

void
pgrel(struct pgrp *pgrp)
{
        procglob_t *prg;
        int count;
        int n;

        n = PGRP_HASH(pgrp->pg_id);
        prg = &procglob[n];

        for (;;) {
                count = pgrp->pg_refs;
                cpu_ccfence();
                KKASSERT(count > 0);
                if (count == 1) {
                        lwkt_gettoken(&prg->proc_token);
                        if (atomic_cmpset_int(&pgrp->pg_refs, 1, 0))
                                break;
                        lwkt_reltoken(&prg->proc_token);
                        /* retry */
                } else {
                        if (atomic_cmpset_int(&pgrp->pg_refs, count, count - 1))
                                return;
                        /* retry */
                }
        }

        /*
         * Successful 1->0 transition, pghash_spin is held.
         */
        LIST_REMOVE(pgrp, pg_list);
        pid_doms[pgrp->pg_id % PIDSEL_DOMAINS] = (uint8_t)time_second;

        /*
         * Reset any sigio structures pointing to us as a result of
         * F_SETOWN with our pgid.
         */
        funsetownlst(&pgrp->pg_sigiolst);

        if (pgrp->pg_session->s_ttyp != NULL &&
            pgrp->pg_session->s_ttyp->t_pgrp == pgrp) {
                pgrp->pg_session->s_ttyp->t_pgrp = NULL;
        }
        lwkt_reltoken(&prg->proc_token);

        sess_rele(pgrp->pg_session);
        kfree(pgrp, M_PGRP);
}

/*
 * Locate a process group by number.  The returned process group will be
 * referenced w/pgref() and must be released with pgrel() (or assigned
 * somewhere if you wish to keep the reference).
 *
 * No requirements.
 */
struct pgrp *
pgfind(pid_t pgid)
{
        struct pgrp *pgrp;
        procglob_t *prg;
        int n;

        n = PGRP_HASH(pgid);
        prg = &procglob[n];
        lwkt_gettoken_shared(&prg->proc_token);

        LIST_FOREACH(pgrp, &prg->allpgrp, pg_list) {
                if (pgrp->pg_id == pgid) {
                        refcount_acquire(&pgrp->pg_refs);
                        lwkt_reltoken(&prg->proc_token);
                        return (pgrp);
                }
        }
        lwkt_reltoken(&prg->proc_token);
        return (NULL);
}

/*
 * Move p to a new or existing process group (and session)
 *
 * No requirements.
 */
int
enterpgrp(struct proc *p, pid_t pgid, int mksess)
{
        struct pgrp *pgrp;
        struct pgrp *opgrp;
        int error;

        pgrp = pgfind(pgid);

        KASSERT(pgrp == NULL || !mksess,
                ("enterpgrp: setsid into non-empty pgrp"));
        KASSERT(!SESS_LEADER(p),
                ("enterpgrp: session leader attempted setpgrp"));

        if (pgrp == NULL) {
                pid_t savepid = p->p_pid;
                struct proc *np;
                procglob_t *prg;
                int n;

                /*
                 * new process group
                 */
                KASSERT(p->p_pid == pgid,
                        ("enterpgrp: new pgrp and pid != pgid"));
                pgrp = kmalloc(sizeof(struct pgrp), M_PGRP, M_WAITOK | M_ZERO);
                pgrp->pg_id = pgid;
                LIST_INIT(&pgrp->pg_members);
                pgrp->pg_jobc = 0;
                SLIST_INIT(&pgrp->pg_sigiolst);
                lwkt_token_init(&pgrp->pg_token, "pgrp_token");
                refcount_init(&pgrp->pg_refs, 1);
                lockinit(&pgrp->pg_lock, "pgwt", 0, 0);

                n = PGRP_HASH(pgid);
                prg = &procglob[n];

                if ((np = pfindn(savepid)) == NULL || np != p) {
                        lwkt_reltoken(&prg->proc_token);
                        error = ESRCH;
                        kfree(pgrp, M_PGRP);
                        goto fatal;
                }

                lwkt_gettoken(&prg->proc_token);
                if (mksess) {
                        struct session *sess;

                        /*
                         * new session
                         */
                        sess = kmalloc(sizeof(struct session), M_SESSION,
                                       M_WAITOK | M_ZERO);
                        lwkt_gettoken(&p->p_token);
                        sess->s_leader = p;
                        sess->s_sid = p->p_pid;
                        sess->s_count = 1;
                        sess->s_ttyvp = NULL;
                        sess->s_ttyp = NULL;
                        bcopy(p->p_session->s_login, sess->s_login,
                              sizeof(sess->s_login));
                        pgrp->pg_session = sess;
                        KASSERT(p == curproc,
                                ("enterpgrp: mksession and p != curproc"));
                        p->p_flags &= ~P_CONTROLT;
                        LIST_INSERT_HEAD(&prg->allsess, sess, s_list);
                        lwkt_reltoken(&p->p_token);
                } else {
                        lwkt_gettoken(&p->p_token);
                        pgrp->pg_session = p->p_session;
                        sess_hold(pgrp->pg_session);
                        lwkt_reltoken(&p->p_token);
                }
                LIST_INSERT_HEAD(&prg->allpgrp, pgrp, pg_list);

                lwkt_reltoken(&prg->proc_token);
        } else if (pgrp == p->p_pgrp) {
                pgrel(pgrp);
                goto done;
        } /* else pgfind() referenced the pgrp */

        lwkt_gettoken(&pgrp->pg_token);
        lwkt_gettoken(&p->p_token);

        /*
         * Replace p->p_pgrp, handling any races that occur.
         */
        while ((opgrp = p->p_pgrp) != NULL) {
                pgref(opgrp);
                lwkt_gettoken(&opgrp->pg_token);
                if (opgrp != p->p_pgrp) {
                        lwkt_reltoken(&opgrp->pg_token);
                        pgrel(opgrp);
                        continue;
                }
                LIST_REMOVE(p, p_pglist);
                break;
        }
        p->p_pgrp = pgrp;
        LIST_INSERT_HEAD(&pgrp->pg_members, p, p_pglist);

        /*
         * Adjust eligibility of affected pgrps to participate in job control.
         * Increment eligibility counts before decrementing, otherwise we
         * could reach 0 spuriously during the first call.
         */
        fixjobc(p, pgrp, 1);
        if (opgrp) {
                fixjobc(p, opgrp, 0);
                lwkt_reltoken(&opgrp->pg_token);
                pgrel(opgrp);   /* manual pgref */
                pgrel(opgrp);   /* p->p_pgrp ref */
        }
        lwkt_reltoken(&p->p_token);
        lwkt_reltoken(&pgrp->pg_token);
done:
        error = 0;
fatal:
        return (error);
}

/*
 * Remove process from process group
 *
 * No requirements.
 */
int
leavepgrp(struct proc *p)
{
        struct pgrp *pg = p->p_pgrp;

        lwkt_gettoken(&p->p_token);
        while ((pg = p->p_pgrp) != NULL) {
                pgref(pg);
                lwkt_gettoken(&pg->pg_token);
                if (p->p_pgrp != pg) {
                        lwkt_reltoken(&pg->pg_token);
                        pgrel(pg);
                        continue;
                }
                p->p_pgrp = NULL;
                LIST_REMOVE(p, p_pglist);
                lwkt_reltoken(&pg->pg_token);
                pgrel(pg);      /* manual pgref */
                pgrel(pg);      /* p->p_pgrp ref */
                break;
        }
        lwkt_reltoken(&p->p_token);

        return (0);
}

/*
 * Adjust the ref count on a session structure.  When the ref count falls to
 * zero the tty is disassociated from the session and the session structure
 * is freed.  Note that tty assocation is not itself ref-counted.
 *
 * No requirements.
 */
void
sess_hold(struct session *sp)
{
        atomic_add_int(&sp->s_count, 1);
}

/*
 * No requirements.
 */
void
sess_rele(struct session *sess)
{
        procglob_t *prg;
        struct tty *tp;
        int count;
        int n;

        n = SESS_HASH(sess->s_sid);
        prg = &procglob[n];

        for (;;) {
                count = sess->s_count;
                cpu_ccfence();
                KKASSERT(count > 0);
                if (count == 1) {
                        lwkt_gettoken(&tty_token);
                        lwkt_gettoken(&prg->proc_token);
                        if (atomic_cmpset_int(&sess->s_count, 1, 0))
                                break;
                        lwkt_reltoken(&prg->proc_token);
                        lwkt_reltoken(&tty_token);
                        /* retry */
                } else {
                        if (atomic_cmpset_int(&sess->s_count, count, count - 1))
                                return;
                        /* retry */
                }
        }

        /*
         * Successful 1->0 transition and tty_token is held.
         */
        LIST_REMOVE(sess, s_list);
        pid_doms[sess->s_sid % PIDSEL_DOMAINS] = (uint8_t)time_second;

        if (sess->s_ttyp && sess->s_ttyp->t_session) {
#ifdef TTY_DO_FULL_CLOSE
                /* FULL CLOSE, see ttyclearsession() */
                KKASSERT(sess->s_ttyp->t_session == sess);
                sess->s_ttyp->t_session = NULL;
#else
                /* HALF CLOSE, see ttyclearsession() */
                if (sess->s_ttyp->t_session == sess)
                        sess->s_ttyp->t_session = NULL;
#endif
        }
        if ((tp = sess->s_ttyp) != NULL) {
                sess->s_ttyp = NULL;
                ttyunhold(tp);
        }
        lwkt_reltoken(&prg->proc_token);
        lwkt_reltoken(&tty_token);

        kfree(sess, M_SESSION);
}

/*
 * Adjust pgrp jobc counters when specified process changes process group.
 * We count the number of processes in each process group that "qualify"
 * the group for terminal job control (those with a parent in a different
 * process group of the same session).  If that count reaches zero, the
 * process group becomes orphaned.  Check both the specified process'
 * process group and that of its children.
 * entering == 0 => p is leaving specified group.
 * entering == 1 => p is entering specified group.
 *
 * No requirements.
 */
void
fixjobc(struct proc *p, struct pgrp *pgrp, int entering)
{
        struct pgrp *hispgrp;
        struct session *mysession;
        struct proc *np;

        /*
         * Check p's parent to see whether p qualifies its own process
         * group; if so, adjust count for p's process group.
         */
        lwkt_gettoken(&p->p_token);     /* p_children scan */
        lwkt_gettoken(&pgrp->pg_token);

        mysession = pgrp->pg_session;
        if ((hispgrp = p->p_pptr->p_pgrp) != pgrp &&
            hispgrp->pg_session == mysession) {
                if (entering)
                        pgrp->pg_jobc++;
                else if (--pgrp->pg_jobc == 0)
                        orphanpg(pgrp);
        }

        /*
         * Check this process' children to see whether they qualify
         * their process groups; if so, adjust counts for children's
         * process groups.
         */
        LIST_FOREACH(np, &p->p_children, p_sibling) {
                PHOLD(np);
                lwkt_gettoken(&np->p_token);
                if ((hispgrp = np->p_pgrp) != pgrp &&
                    hispgrp->pg_session == mysession &&
                    np->p_stat != SZOMB) {
                        pgref(hispgrp);
                        lwkt_gettoken(&hispgrp->pg_token);
                        if (entering)
                                hispgrp->pg_jobc++;
                        else if (--hispgrp->pg_jobc == 0)
                                orphanpg(hispgrp);
                        lwkt_reltoken(&hispgrp->pg_token);
                        pgrel(hispgrp);
                }
                lwkt_reltoken(&np->p_token);
                PRELE(np);
        }
        KKASSERT(pgrp->pg_refs > 0);
        lwkt_reltoken(&pgrp->pg_token);
        lwkt_reltoken(&p->p_token);
}

/*
 * A process group has become orphaned;
 * if there are any stopped processes in the group,
 * hang-up all process in that group.
 *
 * The caller must hold pg_token.
 */
static void
orphanpg(struct pgrp *pg)
{
        struct proc *p;

        LIST_FOREACH(p, &pg->pg_members, p_pglist) {
                if (p->p_stat == SSTOP) {
                        LIST_FOREACH(p, &pg->pg_members, p_pglist) {
                                ksignal(p, SIGHUP);
                                ksignal(p, SIGCONT);
                        }
                        return;
                }
        }
}

/*
 * Add a new process to the allproc list and the PID hash.  This
 * also assigns a pid to the new process.
 *
 * No requirements.
 */
void
proc_add_allproc(struct proc *p)
{
        int random_offset;

        if ((random_offset = randompid) != 0) {
                read_random(&random_offset, sizeof(random_offset));
                random_offset = (random_offset & 0x7FFFFFFF) % randompid;
        }
        proc_makepid(p, random_offset);
}

/*
 * Calculate a new process pid.  This function is integrated into
 * proc_add_allproc() to guarentee that the new pid is not reused before
 * the new process can be added to the allproc list.
 *
 * p_pid is assigned and the process is added to the allproc hash table
 *
 * WARNING! We need to allocate PIDs sequentially during early boot.
 *          In particular, init needs to have a pid of 1.
 */
static
void
proc_makepid(struct proc *p, int random_offset)
{
        static pid_t nextpid = 1;       /* heuristic, allowed to race */
        procglob_t *prg;
        struct pgrp *pg;
        struct proc *ps;
        struct session *sess;
        pid_t base;
        int8_t delta8;
        int retries;
        int n;

        /*
         * Select the next pid base candidate.
         *
         * Check cyclement, do not allow a pid < 100.
         */
        retries = 0;
retry:
        base = atomic_fetchadd_int(&nextpid, 1) + random_offset;
        if (base <= 0 || base >= PID_MAX) {
                base = base % PID_MAX;
                if (base < 0)
                        base = 100;
                if (base < 100)
                        base += 100;
                nextpid = base;         /* reset (SMP race ok) */
        }

        /*
         * Do not allow a base pid to be selected from a domain that has
         * recently seen a pid/pgid/sessid reap.  Sleep a little if we looped
         * through all available domains.
         *
         * WARNING: We want the early pids to be allocated linearly,
         *          particularly pid 1 and pid 2.
         */
        if (++retries >= PIDSEL_DOMAINS)
                tsleep(&nextpid, 0, "makepid", 1);
        if (base >= 100) {
                delta8 = (int8_t)time_second -
                         (int8_t)pid_doms[base % PIDSEL_DOMAINS];
                if (delta8 >= 0 && delta8 <= PIDDOM_DELAY) {
                        ++pid_domain_skips;
                        goto retry;
                }
        }

        /*
         * Calculate a hash index and find an unused process id within
         * the table, looping if we cannot find one.
         *
         * The inner loop increments by ALLPROC_HSIZE which keeps the
         * PID at the same pid_doms[] index as well as the same hash index.
         */
        n = ALLPROC_HASH(base);
        prg = &procglob[n];
        lwkt_gettoken(&prg->proc_token);

restart1:
        LIST_FOREACH(ps, &prg->allproc, p_list) {
                if (ps->p_pid == base) {
                        base += ALLPROC_HSIZE;
                        if (base >= PID_MAX) {
                                lwkt_reltoken(&prg->proc_token);
                                goto retry;
                        }
                        ++pid_inner_skips;
                        goto restart1;
                }
        }
        LIST_FOREACH(pg, &prg->allpgrp, pg_list) {
                if (pg->pg_id == base) {
                        base += ALLPROC_HSIZE;
                        if (base >= PID_MAX) {
                                lwkt_reltoken(&prg->proc_token);
                                goto retry;
                        }
                        ++pid_inner_skips;
                        goto restart1;
                }
        }
        LIST_FOREACH(sess, &prg->allsess, s_list) {
                if (sess->s_sid == base) {
                        base += ALLPROC_HSIZE;
                        if (base >= PID_MAX) {
                                lwkt_reltoken(&prg->proc_token);
                                goto retry;
                        }
                        ++pid_inner_skips;
                        goto restart1;
                }
        }

        /*
         * Assign the pid and insert the process.
         */
        p->p_pid = base;
        LIST_INSERT_HEAD(&prg->allproc, p, p_list);
        lwkt_reltoken(&prg->proc_token);
}

/*
 * Called from exit1 to place the process into a zombie state.
 * The process is removed from the pid hash and p_stat is set
 * to SZOMB.  Normal pfind[n]() calls will not find it any more.
 *
 * Caller must hold p->p_token.  We are required to wait until p_lock
 * becomes zero before we can manipulate the list, allowing allproc
 * scans to guarantee consistency during a list scan.
 */
void
proc_move_allproc_zombie(struct proc *p)
{
        procglob_t *prg;
        int n;

        n = ALLPROC_HASH(p->p_pid);
        prg = &procglob[n];
        PSTALL(p, "reap1", 0);
        lwkt_gettoken(&prg->proc_token);

        PSTALL(p, "reap1a", 0);
        p->p_stat = SZOMB;

        lwkt_reltoken(&prg->proc_token);
        dsched_exit_proc(p);
}

/*
 * This routine is called from kern_wait() and will remove the process
 * from the zombie list and the sibling list.  This routine will block
 * if someone has a lock on the proces (p_lock).
 *
 * Caller must hold p->p_token.  We are required to wait until p_lock
 * becomes zero before we can manipulate the list, allowing allproc
 * scans to guarantee consistency during a list scan.
 */
void
proc_remove_zombie(struct proc *p)
{
        procglob_t *prg;
        int n;

        n = ALLPROC_HASH(p->p_pid);
        prg = &procglob[n];

        PSTALL(p, "reap2", 0);
        lwkt_gettoken(&prg->proc_token);
        PSTALL(p, "reap2a", 0);
        LIST_REMOVE(p, p_list);         /* from remove master list */
        LIST_REMOVE(p, p_sibling);      /* and from sibling list */
        p->p_pptr = NULL;
        pid_doms[p->p_pid % PIDSEL_DOMAINS] = (uint8_t)time_second;
        lwkt_reltoken(&prg->proc_token);
}

/*
 * Handle various requirements prior to returning to usermode.  Called from
 * platform trap and system call code.
 */
void
lwpuserret(struct lwp *lp)
{
        struct proc *p = lp->lwp_proc;

        if (lp->lwp_mpflags & LWP_MP_VNLRU) {
                atomic_clear_int(&lp->lwp_mpflags, LWP_MP_VNLRU);
                allocvnode_gc();
        }
        if (lp->lwp_mpflags & LWP_MP_WEXIT) {
                lwkt_gettoken(&p->p_token);
                lwp_exit(0, NULL);
                lwkt_reltoken(&p->p_token);     /* NOT REACHED */
        }
}

/*
 * Kernel threads run from user processes can also accumulate deferred
 * actions which need to be acted upon.  Callers include:
 *
 * nfsd         - Can allocate lots of vnodes
 */
void
lwpkthreaddeferred(void)
{
        struct lwp *lp = curthread->td_lwp;

        if (lp) {
                if (lp->lwp_mpflags & LWP_MP_VNLRU) {
                        atomic_clear_int(&lp->lwp_mpflags, LWP_MP_VNLRU);
                        allocvnode_gc();
                }
        }
}

void
proc_usermap(struct proc *p, int invfork)
{
        struct sys_upmap *upmap;

        lwkt_gettoken(&p->p_token);
        upmap = kmalloc(roundup2(sizeof(*upmap), PAGE_SIZE), M_PROC,
                        M_WAITOK | M_ZERO);
        if (p->p_upmap == NULL) {
                upmap->header[0].type = UKPTYPE_VERSION;
                upmap->header[0].offset = offsetof(struct sys_upmap, version);
                upmap->header[1].type = UPTYPE_RUNTICKS;
                upmap->header[1].offset = offsetof(struct sys_upmap, runticks);
                upmap->header[2].type = UPTYPE_FORKID;
                upmap->header[2].offset = offsetof(struct sys_upmap, forkid);
                upmap->header[3].type = UPTYPE_PID;
                upmap->header[3].offset = offsetof(struct sys_upmap, pid);
                upmap->header[4].type = UPTYPE_PROC_TITLE;
                upmap->header[4].offset = offsetof(struct sys_upmap,proc_title);
                upmap->header[5].type = UPTYPE_INVFORK;
                upmap->header[5].offset = offsetof(struct sys_upmap, invfork);

                upmap->version = UPMAP_VERSION;
                upmap->pid = p->p_pid;
                upmap->forkid = p->p_forkid;
                upmap->invfork = invfork;
                p->p_upmap = upmap;
        } else {
                kfree(upmap, M_PROC);
        }
        lwkt_reltoken(&p->p_token);
}

void
proc_userunmap(struct proc *p)
{
        struct sys_upmap *upmap;

        lwkt_gettoken(&p->p_token);
        if ((upmap = p->p_upmap) != NULL) {
                p->p_upmap = NULL;
                kfree(upmap, M_PROC);
        }
        lwkt_reltoken(&p->p_token);
}

/*
 * Scan all processes on the allproc list.  The process is automatically
 * held for the callback.  A return value of -1 terminates the loop.
 * Zombie procs are skipped.
 *
 * The callback is made with the process held and proc_token held.
 *
 * We limit the scan to the number of processes as-of the start of
 * the scan so as not to get caught up in an endless loop if new processes
 * are created more quickly than we can scan the old ones.  Add a little
 * slop to try to catch edge cases since nprocs can race.
 *
 * No requirements.
 */
void
allproc_scan(int (*callback)(struct proc *, void *), void *data)
{
        int limit = nprocs + ncpus;
        struct proc *p;
        int r;
        int n;

        /*
         * prg->proc_token protects the allproc list and PHOLD() prevents the
         * process from being removed from the allproc list or the zombproc
         * list.
         */
        for (n = 0; n < ALLPROC_HSIZE; ++n) {
                procglob_t *prg = &procglob[n];
                if (LIST_FIRST(&prg->allproc) == NULL)
                        continue;
                lwkt_gettoken(&prg->proc_token);
                LIST_FOREACH(p, &prg->allproc, p_list) {
                        if (p->p_stat == SZOMB)
                                continue;
                        PHOLD(p);
                        r = callback(p, data);
                        PRELE(p);
                        if (r < 0)
                                break;
                        if (--limit < 0)
                                break;
                }
                lwkt_reltoken(&prg->proc_token);

                /*
                 * Check if asked to stop early
                 */
                if (p)
                        break;
        }
}

/*
 * Scan all lwps of processes on the allproc list.  The lwp is automatically
 * held for the callback.  A return value of -1 terminates the loop.
 *
 * The callback is made with the proces and lwp both held, and proc_token held.
 *
 * No requirements.
 */
void
alllwp_scan(int (*callback)(struct lwp *, void *), void *data)
{
        struct proc *p;
        struct lwp *lp;
        int r = 0;
        int n;

        for (n = 0; n < ALLPROC_HSIZE; ++n) {
                procglob_t *prg = &procglob[n];

                if (LIST_FIRST(&prg->allproc) == NULL)
                        continue;
                lwkt_gettoken(&prg->proc_token);
                LIST_FOREACH(p, &prg->allproc, p_list) {
                        if (p->p_stat == SZOMB)
                                continue;
                        PHOLD(p);
                        lwkt_gettoken(&p->p_token);
                        FOREACH_LWP_IN_PROC(lp, p) {
                                LWPHOLD(lp);
                                r = callback(lp, data);
                                LWPRELE(lp);
                        }
                        lwkt_reltoken(&p->p_token);
                        PRELE(p);
                        if (r < 0)
                                break;
                }
                lwkt_reltoken(&prg->proc_token);

                /*
                 * Asked to exit early
                 */
                if (p)
                        break;
        }
}

/*
 * Scan all processes on the zombproc list.  The process is automatically
 * held for the callback.  A return value of -1 terminates the loop.
 *
 * No requirements.
 * The callback is made with the proces held and proc_token held.
 */
void
zombproc_scan(int (*callback)(struct proc *, void *), void *data)
{
        struct proc *p;
        int r;
        int n;

        /*
         * prg->proc_token protects the allproc list and PHOLD() prevents the
         * process from being removed from the allproc list or the zombproc
         * list.
         */
        for (n = 0; n < ALLPROC_HSIZE; ++n) {
                procglob_t *prg = &procglob[n];

                if (LIST_FIRST(&prg->allproc) == NULL)
                        continue;
                lwkt_gettoken(&prg->proc_token);
                LIST_FOREACH(p, &prg->allproc, p_list) {
                        if (p->p_stat != SZOMB)
                                continue;
                        PHOLD(p);
                        r = callback(p, data);
                        PRELE(p);
                        if (r < 0)
                                break;
                }
                lwkt_reltoken(&prg->proc_token);

                /*
                 * Check if asked to stop early
                 */
                if (p)
                        break;
        }
}

#include "opt_ddb.h"
#ifdef DDB
#include <ddb/ddb.h>

/*
 * Debugging only
 */
DB_SHOW_COMMAND(pgrpdump, pgrpdump)
{
        struct pgrp *pgrp;
        struct proc *p;
        procglob_t *prg;
        int i;

        for (i = 0; i < ALLPROC_HSIZE; ++i) {
                prg = &procglob[i];

                if (LIST_EMPTY(&prg->allpgrp))
                        continue;
                kprintf("\tindx %d\n", i);
                LIST_FOREACH(pgrp, &prg->allpgrp, pg_list) {
                        kprintf("\tpgrp %p, pgid %ld, sess %p, "
                                "sesscnt %d, mem %p\n",
                                (void *)pgrp, (long)pgrp->pg_id,
                                (void *)pgrp->pg_session,
                                pgrp->pg_session->s_count,
                                (void *)LIST_FIRST(&pgrp->pg_members));
                        LIST_FOREACH(p, &pgrp->pg_members, p_pglist) {
                                kprintf("\t\tpid %ld addr %p pgrp %p\n",
                                        (long)p->p_pid, (void *)p,
                                        (void *)p->p_pgrp);
                        }
                }
        }
}
#endif /* DDB */

/*
 * The caller must hold proc_token.
 */
static int
sysctl_out_proc(struct proc *p, struct sysctl_req *req, int flags)
{
        struct kinfo_proc ki;
        struct lwp *lp;
        int skp = 0, had_output = 0;
        int error;

        bzero(&ki, sizeof(ki));
        lwkt_gettoken_shared(&p->p_token);
        fill_kinfo_proc(p, &ki);
        if ((flags & KERN_PROC_FLAG_LWP) == 0)
                skp = 1;
        error = 0;
        FOREACH_LWP_IN_PROC(lp, p) {
                LWPHOLD(lp);
                fill_kinfo_lwp(lp, &ki.kp_lwp);
                had_output = 1;
                error = SYSCTL_OUT(req, &ki, sizeof(ki));
                LWPRELE(lp);
                if (error)
                        break;
                if (skp)
                        break;
        }
        lwkt_reltoken(&p->p_token);
        /* We need to output at least the proc, even if there is no lwp. */
        if (had_output == 0) {
                error = SYSCTL_OUT(req, &ki, sizeof(ki));
        }
        return (error);
}

/*
 * The caller must hold proc_token.
 */
static int
sysctl_out_proc_kthread(struct thread *td, struct sysctl_req *req)
{
        struct kinfo_proc ki;
        int error;

        fill_kinfo_proc_kthread(td, &ki);
        error = SYSCTL_OUT(req, &ki, sizeof(ki));
        if (error)
                return error;
        return(0);
}

/*
 * No requirements.
 */
static int
sysctl_kern_proc(SYSCTL_HANDLER_ARGS)
{
        int *name = (int *)arg1;
        int oid = oidp->oid_number;
        u_int namelen = arg2;
        struct proc *p;
        struct thread *td;
        struct thread *marker;
        int flags = 0;
        int error = 0;
        int n;
        int origcpu;
        struct ucred *cr1 = curproc->p_ucred;

        flags = oid & KERN_PROC_FLAGMASK;
        oid &= ~KERN_PROC_FLAGMASK;

        if ((oid == KERN_PROC_ALL && namelen != 0) ||
            (oid != KERN_PROC_ALL && namelen != 1)) {
                return (EINVAL);
        }

        /*
         * proc_token protects the allproc list and PHOLD() prevents the
         * process from being removed from the allproc list or the zombproc
         * list.
         */
        if (oid == KERN_PROC_PID) {
                p = pfind((pid_t)name[0]);
                if (p) {
                        if (PRISON_CHECK(cr1, p->p_ucred))
                                error = sysctl_out_proc(p, req, flags);
                        PRELE(p);
                }
                goto post_threads;
        }
        p = NULL;

        if (!req->oldptr) {
                /* overestimate by 5 procs */
                error = SYSCTL_OUT(req, 0, sizeof (struct kinfo_proc) * 5);
                if (error)
                        goto post_threads;
        }

        for (n = 0; n < ALLPROC_HSIZE; ++n) {
                procglob_t *prg = &procglob[n];

                if (LIST_EMPTY(&prg->allproc))
                        continue;
                lwkt_gettoken_shared(&prg->proc_token);
                LIST_FOREACH(p, &prg->allproc, p_list) {
                        /*
                         * Show a user only their processes.
                         */
                        if ((!ps_showallprocs) &&
                                (p->p_ucred == NULL || p_trespass(cr1, p->p_ucred))) {
                                continue;
                        }
                        /*
                         * Skip embryonic processes.
                         */
                        if (p->p_stat == SIDL)
                                continue;
                        /*
                         * TODO - make more efficient (see notes below).
                         * do by session.
                         */
                        switch (oid) {
                        case KERN_PROC_PGRP:
                                /* could do this by traversing pgrp */
                                if (p->p_pgrp == NULL || 
                                    p->p_pgrp->pg_id != (pid_t)name[0])
                                        continue;
                                break;

                        case KERN_PROC_TTY:
                                if ((p->p_flags & P_CONTROLT) == 0 ||
                                    p->p_session == NULL ||
                                    p->p_session->s_ttyp == NULL ||
                                    dev2udev(p->p_session->s_ttyp->t_dev) != 
                                        (udev_t)name[0])
                                        continue;
                                break;

                        case KERN_PROC_UID:
                                if (p->p_ucred == NULL || 
                                    p->p_ucred->cr_uid != (uid_t)name[0])
                                        continue;
                                break;

                        case KERN_PROC_RUID:
                                if (p->p_ucred == NULL || 
                                    p->p_ucred->cr_ruid != (uid_t)name[0])
                                        continue;
                                break;
                        }

                        if (!PRISON_CHECK(cr1, p->p_ucred))
                                continue;
                        PHOLD(p);
                        error = sysctl_out_proc(p, req, flags);
                        PRELE(p);
                        if (error) {
                                lwkt_reltoken(&prg->proc_token);
                                goto post_threads;
                        }
                }
                lwkt_reltoken(&prg->proc_token);
        }

        /*
         * Iterate over all active cpus and scan their thread list.  Start
         * with the next logical cpu and end with our original cpu.  We
         * migrate our own thread to each target cpu in order to safely scan
         * its thread list.  In the last loop we migrate back to our original
         * cpu.
         */
        origcpu = mycpu->gd_cpuid;
        if (!ps_showallthreads || jailed(cr1))
                goto post_threads;

        marker = kmalloc(sizeof(struct thread), M_TEMP, M_WAITOK|M_ZERO);
        marker->td_flags = TDF_MARKER;
        error = 0;

        for (n = 1; n <= ncpus; ++n) {
                globaldata_t rgd;
                int nid;

                nid = (origcpu + n) % ncpus;
                if (CPUMASK_TESTBIT(smp_active_mask, nid) == 0)
                        continue;
                rgd = globaldata_find(nid);
                lwkt_setcpu_self(rgd);

                crit_enter();
                TAILQ_INSERT_TAIL(&rgd->gd_tdallq, marker, td_allq);

                while ((td = TAILQ_PREV(marker, lwkt_queue, td_allq)) != NULL) {
                        TAILQ_REMOVE(&rgd->gd_tdallq, marker, td_allq);
                        TAILQ_INSERT_BEFORE(td, marker, td_allq);
                        if (td->td_flags & TDF_MARKER)
                                continue;
                        if (td->td_proc)
                                continue;

                        lwkt_hold(td);
                        crit_exit();

                        switch (oid) {
                        case KERN_PROC_PGRP:
                        case KERN_PROC_TTY:
                        case KERN_PROC_UID:
                        case KERN_PROC_RUID:
                                break;
                        default:
                                error = sysctl_out_proc_kthread(td, req);
                                break;
                        }
                        lwkt_rele(td);
                        crit_enter();
                        if (error)
                                break;
                }
                TAILQ_REMOVE(&rgd->gd_tdallq, marker, td_allq);
                crit_exit();

                if (error)
                        break;
        }

        /*
         * Userland scheduler expects us to return on the same cpu we
         * started on.
         */
        if (mycpu->gd_cpuid != origcpu)
                lwkt_setcpu_self(globaldata_find(origcpu));

        kfree(marker, M_TEMP);

post_threads:
        return (error);
}

/*
 * This sysctl allows a process to retrieve the argument list or process
 * title for another process without groping around in the address space
 * of the other process.  It also allow a process to set its own "process 
 * title to a string of its own choice.
 *
 * No requirements.
 */
static int
sysctl_kern_proc_args(SYSCTL_HANDLER_ARGS)
{
        int *name = (int*) arg1;
        u_int namelen = arg2;
        struct proc *p;
        struct pargs *opa;
        struct pargs *pa;
        int error = 0;
        struct ucred *cr1 = curproc->p_ucred;

        if (namelen != 1) 
                return (EINVAL);

        p = pfind((pid_t)name[0]);
        if (p == NULL)
                goto done;
        lwkt_gettoken(&p->p_token);

        if ((!ps_argsopen) && p_trespass(cr1, p->p_ucred))
                goto done;

        if (req->newptr && curproc != p) {
                error = EPERM;
                goto done;
        }
        if (req->oldptr) {
                if (p->p_upmap != NULL && p->p_upmap->proc_title[0]) {
                        /*
                         * Args set via writable user process mmap.
                         * We must calculate the string length manually
                         * because the user data can change at any time.
                         */
                        size_t n;
                        char *base;

                        base = p->p_upmap->proc_title;
                        for (n = 0; n < UPMAP_MAXPROCTITLE - 1; ++n) {
                                if (base[n] == 0)
                                        break;
                        }
                        error = SYSCTL_OUT(req, base, n);
                        if (error == 0)
                                error = SYSCTL_OUT(req, "", 1);
                } else if ((pa = p->p_args) != NULL) {
                        /*
                         * Args set by setproctitle() sysctl.
                         */
                        refcount_acquire(&pa->ar_ref);
                        error = SYSCTL_OUT(req, pa->ar_args, pa->ar_length);
                        if (refcount_release(&pa->ar_ref))
                                kfree(pa, M_PARGS);
                }
        }
        if (req->newptr == NULL)
                goto done;

        if (req->newlen + sizeof(struct pargs) > ps_arg_cache_limit) {
                goto done;
        }

        pa = kmalloc(sizeof(struct pargs) + req->newlen, M_PARGS, M_WAITOK);
        refcount_init(&pa->ar_ref, 1);
        pa->ar_length = req->newlen;
        error = SYSCTL_IN(req, pa->ar_args, req->newlen);
        if (error) {
                kfree(pa, M_PARGS);
                goto done;
        }


        /*
         * Replace p_args with the new pa.  p_args may have previously
         * been NULL.
         */
        opa = p->p_args;
        p->p_args = pa;

        if (opa) {
                KKASSERT(opa->ar_ref > 0);
                if (refcount_release(&opa->ar_ref)) {
                        kfree(opa, M_PARGS);
                        /* opa = NULL; */
                }
        }
done:
        if (p) {
                lwkt_reltoken(&p->p_token);
                PRELE(p);
        }
        return (error);
}

static int
sysctl_kern_proc_cwd(SYSCTL_HANDLER_ARGS)
{
        int *name = (int*) arg1;
        u_int namelen = arg2;
        struct proc *p;
        int error = 0;
        char *fullpath, *freepath;
        struct ucred *cr1 = curproc->p_ucred;

        if (namelen != 1) 
                return (EINVAL);

        p = pfind((pid_t)name[0]);
        if (p == NULL)
                goto done;
        lwkt_gettoken_shared(&p->p_token);

        /*
         * If we are not allowed to see other args, we certainly shouldn't
         * get the cwd either. Also check the usual trespassing.
         */
        if ((!ps_argsopen) && p_trespass(cr1, p->p_ucred))
                goto done;

        if (req->oldptr && p->p_fd != NULL && p->p_fd->fd_ncdir.ncp) {
                struct nchandle nch;

                cache_copy(&p->p_fd->fd_ncdir, &nch);
                error = cache_fullpath(p, &nch, NULL,
                                       &fullpath, &freepath, 0);
                cache_drop(&nch);
                if (error)
                        goto done;
                error = SYSCTL_OUT(req, fullpath, strlen(fullpath) + 1);
                kfree(freepath, M_TEMP);
        }

done:
        if (p) {
                lwkt_reltoken(&p->p_token);
                PRELE(p);
        }
        return (error);
}

/*
 * This sysctl allows a process to retrieve the path of the executable for
 * itself or another process.
 */
static int
sysctl_kern_proc_pathname(SYSCTL_HANDLER_ARGS)
{
        pid_t *pidp = (pid_t *)arg1;
        unsigned int arglen = arg2;
        struct proc *p;
        char *retbuf, *freebuf;
        int error = 0;
        struct nchandle nch;

        if (arglen != 1)
                return (EINVAL);
        if (*pidp == -1) {      /* -1 means this process */
                p = curproc;
        } else {
                p = pfind(*pidp);
                if (p == NULL)
                        return (ESRCH);
        }

        cache_copy(&p->p_textnch, &nch);
        error = cache_fullpath(p, &nch, NULL, &retbuf, &freebuf, 0);
        cache_drop(&nch);
        if (error)
                goto done;
        error = SYSCTL_OUT(req, retbuf, strlen(retbuf) + 1);
        kfree(freebuf, M_TEMP);
done:
        if (*pidp != -1)
                PRELE(p);

        return (error);
}

SYSCTL_NODE(_kern, KERN_PROC, proc, CTLFLAG_RD,  0, "Process table");

SYSCTL_PROC(_kern_proc, KERN_PROC_ALL, all, CTLFLAG_RD|CTLTYPE_STRUCT,
        0, 0, sysctl_kern_proc, "S,proc", "Return entire process table");

SYSCTL_NODE(_kern_proc, KERN_PROC_PGRP, pgrp, CTLFLAG_RD, 
        sysctl_kern_proc, "Process table");

SYSCTL_NODE(_kern_proc, KERN_PROC_TTY, tty, CTLFLAG_RD, 
        sysctl_kern_proc, "Process table");

SYSCTL_NODE(_kern_proc, KERN_PROC_UID, uid, CTLFLAG_RD, 
        sysctl_kern_proc, "Process table");

SYSCTL_NODE(_kern_proc, KERN_PROC_RUID, ruid, CTLFLAG_RD, 
        sysctl_kern_proc, "Process table");

SYSCTL_NODE(_kern_proc, KERN_PROC_PID, pid, CTLFLAG_RD, 
        sysctl_kern_proc, "Process table");

SYSCTL_NODE(_kern_proc, (KERN_PROC_ALL | KERN_PROC_FLAG_LWP), all_lwp, CTLFLAG_RD,
        sysctl_kern_proc, "Process table");

SYSCTL_NODE(_kern_proc, (KERN_PROC_PGRP | KERN_PROC_FLAG_LWP), pgrp_lwp, CTLFLAG_RD, 
        sysctl_kern_proc, "Process table");

SYSCTL_NODE(_kern_proc, (KERN_PROC_TTY | KERN_PROC_FLAG_LWP), tty_lwp, CTLFLAG_RD, 
        sysctl_kern_proc, "Process table");

SYSCTL_NODE(_kern_proc, (KERN_PROC_UID | KERN_PROC_FLAG_LWP), uid_lwp, CTLFLAG_RD, 
        sysctl_kern_proc, "Process table");

SYSCTL_NODE(_kern_proc, (KERN_PROC_RUID | KERN_PROC_FLAG_LWP), ruid_lwp, CTLFLAG_RD, 
        sysctl_kern_proc, "Process table");

SYSCTL_NODE(_kern_proc, (KERN_PROC_PID | KERN_PROC_FLAG_LWP), pid_lwp, CTLFLAG_RD, 
        sysctl_kern_proc, "Process table");

SYSCTL_NODE(_kern_proc, KERN_PROC_ARGS, args, CTLFLAG_RW | CTLFLAG_ANYBODY,
        sysctl_kern_proc_args, "Process argument list");

SYSCTL_NODE(_kern_proc, KERN_PROC_CWD, cwd, CTLFLAG_RD | CTLFLAG_ANYBODY,
        sysctl_kern_proc_cwd, "Process argument list");

static SYSCTL_NODE(_kern_proc, KERN_PROC_PATHNAME, pathname, CTLFLAG_RD,
        sysctl_kern_proc_pathname, "Process executable path");