More LWKT messaging cleanups. Isolate the default port functions by making

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

/*
 * Copyright (c) 2003 Matthew Dillon <dillon@backplane.com>
 * 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.
 *
 * THIS SOFTWARE IS PROVIDED BY THE AUTHOR 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 AUTHOR 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.
 *
 * $DragonFly: src/sys/kern/lwkt_thread.c,v 1.43 2003/11/24 20:46:01 dillon Exp $
 */

/*
 * Each cpu in a system has its own self-contained light weight kernel
 * thread scheduler, which means that generally speaking we only need
 * to use a critical section to avoid problems.  Foreign thread 
 * scheduling is queued via (async) IPIs.
 *
 * NOTE: on UP machines smp_active is defined to be 0.  On SMP machines
 * smp_active is 0 prior to SMP activation, then it is 1.  The LWKT module
 * uses smp_active to optimize UP builds and to avoid sending IPIs during
 * early boot (primarily interrupt and network thread initialization).
 */

#ifdef _KERNEL

#include <sys/param.h>
#include <sys/systm.h>
#include <sys/kernel.h>
#include <sys/proc.h>
#include <sys/rtprio.h>
#include <sys/queue.h>
#include <sys/thread2.h>
#include <sys/sysctl.h>
#include <sys/kthread.h>
#include <machine/cpu.h>
#include <sys/lock.h>

#include <vm/vm.h>
#include <vm/vm_param.h>
#include <vm/vm_kern.h>
#include <vm/vm_object.h>
#include <vm/vm_page.h>
#include <vm/vm_map.h>
#include <vm/vm_pager.h>
#include <vm/vm_extern.h>
#include <vm/vm_zone.h>

#include <machine/stdarg.h>
#include <machine/ipl.h>
#include <machine/smp.h>

#define THREAD_STACK    (UPAGES * PAGE_SIZE)

#else

#include <sys/stdint.h>
#include <liblwkt/thread.h>
#include <sys/thread.h>
#include <sys/msgport.h>
#include <sys/errno.h>
#include <liblwkt/globaldata.h>
#include <sys/thread2.h>
#include <sys/msgport2.h>
#include <stdlib.h>
#include <machine/cpufunc.h>

#endif

static int untimely_switch = 0;
#ifdef INVARIANTS
static int token_debug = 0;
#endif
static __int64_t switch_count = 0;
static __int64_t preempt_hit = 0;
static __int64_t preempt_miss = 0;
static __int64_t preempt_weird = 0;
static __int64_t ipiq_count = 0;
static __int64_t ipiq_fifofull = 0;

#ifdef _KERNEL

SYSCTL_INT(_lwkt, OID_AUTO, untimely_switch, CTLFLAG_RW, &untimely_switch, 0, "");
#ifdef INVARIANTS
SYSCTL_INT(_lwkt, OID_AUTO, token_debug, CTLFLAG_RW, &token_debug, 0, "");
#endif
SYSCTL_QUAD(_lwkt, OID_AUTO, switch_count, CTLFLAG_RW, &switch_count, 0, "");
SYSCTL_QUAD(_lwkt, OID_AUTO, preempt_hit, CTLFLAG_RW, &preempt_hit, 0, "");
SYSCTL_QUAD(_lwkt, OID_AUTO, preempt_miss, CTLFLAG_RW, &preempt_miss, 0, "");
SYSCTL_QUAD(_lwkt, OID_AUTO, preempt_weird, CTLFLAG_RW, &preempt_weird, 0, "");
SYSCTL_QUAD(_lwkt, OID_AUTO, ipiq_count, CTLFLAG_RW, &ipiq_count, 0, "");
SYSCTL_QUAD(_lwkt, OID_AUTO, ipiq_fifofull, CTLFLAG_RW, &ipiq_fifofull, 0, "");

#endif

/*
 * These helper procedures handle the runq, they can only be called from
 * within a critical section.
 *
 * WARNING!  Prior to SMP being brought up it is possible to enqueue and
 * dequeue threads belonging to other cpus, so be sure to use td->td_gd
 * instead of 'mycpu' when referencing the globaldata structure.   Once
 * SMP live enqueuing and dequeueing only occurs on the current cpu.
 */
static __inline
void
_lwkt_dequeue(thread_t td)
{
    if (td->td_flags & TDF_RUNQ) {
        int nq = td->td_pri & TDPRI_MASK;
        struct globaldata *gd = td->td_gd;

        td->td_flags &= ~TDF_RUNQ;
        TAILQ_REMOVE(&gd->gd_tdrunq[nq], td, td_threadq);
        /* runqmask is passively cleaned up by the switcher */
    }
}

static __inline
void
_lwkt_enqueue(thread_t td)
{
    if ((td->td_flags & TDF_RUNQ) == 0) {
        int nq = td->td_pri & TDPRI_MASK;
        struct globaldata *gd = td->td_gd;

        td->td_flags |= TDF_RUNQ;
        TAILQ_INSERT_TAIL(&gd->gd_tdrunq[nq], td, td_threadq);
        gd->gd_runqmask |= 1 << nq;
    }
}

static __inline
int
_lwkt_wantresched(thread_t ntd, thread_t cur)
{
    return((ntd->td_pri & TDPRI_MASK) > (cur->td_pri & TDPRI_MASK));
}

/*
 * LWKTs operate on a per-cpu basis
 *
 * WARNING!  Called from early boot, 'mycpu' may not work yet.
 */
void
lwkt_gdinit(struct globaldata *gd)
{
    int i;

    for (i = 0; i < sizeof(gd->gd_tdrunq)/sizeof(gd->gd_tdrunq[0]); ++i)
        TAILQ_INIT(&gd->gd_tdrunq[i]);
    gd->gd_runqmask = 0;
    TAILQ_INIT(&gd->gd_tdallq);
}

/*
 * Initialize a thread wait structure prior to first use.
 *
 * NOTE!  called from low level boot code, we cannot do anything fancy!
 */
void
lwkt_init_wait(lwkt_wait_t w)
{
    TAILQ_INIT(&w->wa_waitq);
}

/*
 * Create a new thread.  The thread must be associated with a process context
 * or LWKT start address before it can be scheduled.  If the target cpu is
 * -1 the thread will be created on the current cpu.
 *
 * If you intend to create a thread without a process context this function
 * does everything except load the startup and switcher function.
 */
thread_t
lwkt_alloc_thread(struct thread *td, int cpu)
{
    void *stack;
    int flags = 0;

    if (td == NULL) {
        crit_enter();
        if (mycpu->gd_tdfreecount > 0) {
            --mycpu->gd_tdfreecount;
            td = TAILQ_FIRST(&mycpu->gd_tdfreeq);
            KASSERT(td != NULL && (td->td_flags & TDF_RUNNING) == 0,
                ("lwkt_alloc_thread: unexpected NULL or corrupted td"));
            TAILQ_REMOVE(&mycpu->gd_tdfreeq, td, td_threadq);
            crit_exit();
            stack = td->td_kstack;
            flags = td->td_flags & (TDF_ALLOCATED_STACK|TDF_ALLOCATED_THREAD);
        } else {
            crit_exit();
#ifdef _KERNEL
            td = zalloc(thread_zone);
#else
            td = malloc(sizeof(struct thread));
#endif
            td->td_kstack = NULL;
            flags |= TDF_ALLOCATED_THREAD;
        }
    }
    if ((stack = td->td_kstack) == NULL) {
#ifdef _KERNEL
        stack = (void *)kmem_alloc(kernel_map, THREAD_STACK);
#else
        stack = liblwkt_alloc_stack(THREAD_STACK);
#endif
        flags |= TDF_ALLOCATED_STACK;
    }
    if (cpu < 0)
        lwkt_init_thread(td, stack, flags, mycpu);
    else
        lwkt_init_thread(td, stack, flags, globaldata_find(cpu));
    return(td);
}

/*
 * Initialize a preexisting thread structure.  This function is used by
 * lwkt_alloc_thread() and also used to initialize the per-cpu idlethread.
 *
 * All threads start out in a critical section at a priority of
 * TDPRI_KERN_DAEMON.  Higher level code will modify the priority as
 * appropriate.  This function may send an IPI message when the 
 * requested cpu is not the current cpu and consequently gd_tdallq may
 * not be initialized synchronously from the point of view of the originating
 * cpu.
 *
 * NOTE! we have to be careful in regards to creating threads for other cpus
 * if SMP has not yet been activated.
 */
static void
lwkt_init_thread_remote(void *arg)
{
    thread_t td = arg;

    TAILQ_INSERT_TAIL(&td->td_gd->gd_tdallq, td, td_allq);
}

void
lwkt_init_thread(thread_t td, void *stack, int flags, struct globaldata *gd)
{
    bzero(td, sizeof(struct thread));
    td->td_kstack = stack;
    td->td_flags |= flags;
    td->td_gd = gd;
    td->td_pri = TDPRI_KERN_DAEMON + TDPRI_CRIT;
    lwkt_initport(&td->td_msgport, td);
    pmap_init_thread(td);
    if (smp_active == 0 || gd == mycpu) {
        crit_enter();
        TAILQ_INSERT_TAIL(&gd->gd_tdallq, td, td_allq);
        crit_exit();
    } else {
        lwkt_send_ipiq(gd->gd_cpuid, lwkt_init_thread_remote, td);
    }
}

void
lwkt_set_comm(thread_t td, const char *ctl, ...)
{
    __va_list va;

    __va_start(va, ctl);
    vsnprintf(td->td_comm, sizeof(td->td_comm), ctl, va);
    __va_end(va);
}

void
lwkt_hold(thread_t td)
{
    ++td->td_refs;
}

void
lwkt_rele(thread_t td)
{
    KKASSERT(td->td_refs > 0);
    --td->td_refs;
}

#ifdef _KERNEL

void
lwkt_wait_free(thread_t td)
{
    while (td->td_refs)
        tsleep(td, 0, "tdreap", hz);
}

#endif

void
lwkt_free_thread(thread_t td)
{
    struct globaldata *gd = mycpu;

    KASSERT((td->td_flags & TDF_RUNNING) == 0,
        ("lwkt_free_thread: did not exit! %p", td));

    crit_enter();
    TAILQ_REMOVE(&gd->gd_tdallq, td, td_allq);
    if (gd->gd_tdfreecount < CACHE_NTHREADS &&
        (td->td_flags & TDF_ALLOCATED_THREAD)
    ) {
        ++gd->gd_tdfreecount;
        TAILQ_INSERT_HEAD(&gd->gd_tdfreeq, td, td_threadq);
        crit_exit();
    } else {
        crit_exit();
        if (td->td_kstack && (td->td_flags & TDF_ALLOCATED_STACK)) {
#ifdef _KERNEL
            kmem_free(kernel_map, (vm_offset_t)td->td_kstack, THREAD_STACK);
#else
            liblwkt_free_stack(td->td_kstack, THREAD_STACK);
#endif
            /* gd invalid */
            td->td_kstack = NULL;
        }
        if (td->td_flags & TDF_ALLOCATED_THREAD) {
#ifdef _KERNEL
            zfree(thread_zone, td);
#else
            free(td);
#endif
        }
    }
}


/*
 * Switch to the next runnable lwkt.  If no LWKTs are runnable then 
 * switch to the idlethread.  Switching must occur within a critical
 * section to avoid races with the scheduling queue.
 *
 * We always have full control over our cpu's run queue.  Other cpus
 * that wish to manipulate our queue must use the cpu_*msg() calls to
 * talk to our cpu, so a critical section is all that is needed and
 * the result is very, very fast thread switching.
 *
 * The LWKT scheduler uses a fixed priority model and round-robins at
 * each priority level.  User process scheduling is a totally
 * different beast and LWKT priorities should not be confused with
 * user process priorities.
 *
 * The MP lock may be out of sync with the thread's td_mpcount.  lwkt_switch()
 * cleans it up.  Note that the td_switch() function cannot do anything that
 * requires the MP lock since the MP lock will have already been setup for
 * the target thread (not the current thread).  It's nice to have a scheduler
 * that does not need the MP lock to work because it allows us to do some
 * really cool high-performance MP lock optimizations.
 */

void
lwkt_switch(void)
{
    struct globaldata *gd;
    thread_t td = curthread;
    thread_t ntd;
#ifdef SMP
    int mpheld;
#endif

    /*
     * Switching from within a 'fast' (non thread switched) interrupt is
     * illegal.
     */
    if (mycpu->gd_intr_nesting_level && panicstr == NULL) {
        panic("lwkt_switch: cannot switch from within a fast interrupt, yet\n");
    }

    /*
     * Passive release (used to transition from user to kernel mode
     * when we block or switch rather then when we enter the kernel).
     * This function is NOT called if we are switching into a preemption
     * or returning from a preemption.  Typically this causes us to lose
     * our P_CURPROC designation (if we have one) and become a true LWKT
     * thread, and may also hand P_CURPROC to another process and schedule
     * its thread.
     */
    if (td->td_release)
            td->td_release(td);

    crit_enter();
    ++switch_count;

#ifdef SMP
    /*
     * td_mpcount cannot be used to determine if we currently hold the
     * MP lock because get_mplock() will increment it prior to attempting
     * to get the lock, and switch out if it can't.  Our ownership of 
     * the actual lock will remain stable while we are in a critical section
     * (but, of course, another cpu may own or release the lock so the
     * actual value of mp_lock is not stable).
     */
    mpheld = MP_LOCK_HELD();
#endif
    if ((ntd = td->td_preempted) != NULL) {
        /*
         * We had preempted another thread on this cpu, resume the preempted
         * thread.  This occurs transparently, whether the preempted thread
         * was scheduled or not (it may have been preempted after descheduling
         * itself). 
         *
         * We have to setup the MP lock for the original thread after backing
         * out the adjustment that was made to curthread when the original
         * was preempted.
         */
        KKASSERT(ntd->td_flags & TDF_PREEMPT_LOCK);
#ifdef SMP
        if (ntd->td_mpcount && mpheld == 0) {
            panic("MPLOCK NOT HELD ON RETURN: %p %p %d %d\n",
               td, ntd, td->td_mpcount, ntd->td_mpcount);
        }
        if (ntd->td_mpcount) {
            td->td_mpcount -= ntd->td_mpcount;
            KKASSERT(td->td_mpcount >= 0);
        }
#endif
        ntd->td_flags |= TDF_PREEMPT_DONE;
        /* YYY release mp lock on switchback if original doesn't need it */
    } else {
        /*
         * Priority queue / round-robin at each priority.  Note that user
         * processes run at a fixed, low priority and the user process
         * scheduler deals with interactions between user processes
         * by scheduling and descheduling them from the LWKT queue as
         * necessary.
         *
         * We have to adjust the MP lock for the target thread.  If we 
         * need the MP lock and cannot obtain it we try to locate a
         * thread that does not need the MP lock.
         */
        gd = mycpu;
again:
        if (gd->gd_runqmask) {
            int nq = bsrl(gd->gd_runqmask);
            if ((ntd = TAILQ_FIRST(&gd->gd_tdrunq[nq])) == NULL) {
                gd->gd_runqmask &= ~(1 << nq);
                goto again;
            }
#ifdef SMP
            if (ntd->td_mpcount && mpheld == 0 && !cpu_try_mplock()) {
                /*
                 * Target needs MP lock and we couldn't get it, try
                 * to locate a thread which does not need the MP lock
                 * to run.  If we cannot locate a thread spin in idle.
                 */
                u_int32_t rqmask = gd->gd_runqmask;
                while (rqmask) {
                    TAILQ_FOREACH(ntd, &gd->gd_tdrunq[nq], td_threadq) {
                        if (ntd->td_mpcount == 0)
                            break;
                    }
                    if (ntd)
                        break;
                    rqmask &= ~(1 << nq);
                    nq = bsrl(rqmask);
                }
                if (ntd == NULL) {
                    ntd = &gd->gd_idlethread;
                    ntd->td_flags |= TDF_IDLE_NOHLT;
                } else {
                    TAILQ_REMOVE(&gd->gd_tdrunq[nq], ntd, td_threadq);
                    TAILQ_INSERT_TAIL(&gd->gd_tdrunq[nq], ntd, td_threadq);
                }
            } else {
                TAILQ_REMOVE(&gd->gd_tdrunq[nq], ntd, td_threadq);
                TAILQ_INSERT_TAIL(&gd->gd_tdrunq[nq], ntd, td_threadq);
            }
#else
            TAILQ_REMOVE(&gd->gd_tdrunq[nq], ntd, td_threadq);
            TAILQ_INSERT_TAIL(&gd->gd_tdrunq[nq], ntd, td_threadq);
#endif
        } else {
            /*
             * Nothing to run but we may still need the BGL to deal with
             * pending interrupts, spin in idle if so.
             */
            ntd = &gd->gd_idlethread;
            if (gd->gd_reqflags)
                ntd->td_flags |= TDF_IDLE_NOHLT;
        }
    }
    KASSERT(ntd->td_pri >= TDPRI_CRIT,
        ("priority problem in lwkt_switch %d %d", td->td_pri, ntd->td_pri));

    /*
     * Do the actual switch.  If the new target does not need the MP lock
     * and we are holding it, release the MP lock.  If the new target requires
     * the MP lock we have already acquired it for the target.
     */
#ifdef SMP
    if (ntd->td_mpcount == 0 ) {
        if (MP_LOCK_HELD())
            cpu_rel_mplock();
    } else {
        ASSERT_MP_LOCK_HELD();
    }
#endif
    if (td != ntd) {
        td->td_switch(ntd);
    }

    crit_exit();
}

/*
 * Switch if another thread has a higher priority.  Do not switch to other
 * threads at the same priority.
 */
void
lwkt_maybe_switch()
{
    struct globaldata *gd = mycpu;
    struct thread *td = gd->gd_curthread;

    if ((td->td_pri & TDPRI_MASK) < bsrl(gd->gd_runqmask)) {
        lwkt_switch();
    }
}

/*
 * Request that the target thread preempt the current thread.  Preemption
 * only works under a specific set of conditions:
 *
 *      - We are not preempting ourselves
 *      - The target thread is owned by the current cpu
 *      - We are not currently being preempted
 *      - The target is not currently being preempted
 *      - We are able to satisfy the target's MP lock requirements (if any).
 *
 * THE CALLER OF LWKT_PREEMPT() MUST BE IN A CRITICAL SECTION.  Typically
 * this is called via lwkt_schedule() through the td_preemptable callback.
 * critpri is the managed critical priority that we should ignore in order
 * to determine whether preemption is possible (aka usually just the crit
 * priority of lwkt_schedule() itself).
 *
 * XXX at the moment we run the target thread in a critical section during
 * the preemption in order to prevent the target from taking interrupts
 * that *WE* can't.  Preemption is strictly limited to interrupt threads
 * and interrupt-like threads, outside of a critical section, and the
 * preempted source thread will be resumed the instant the target blocks
 * whether or not the source is scheduled (i.e. preemption is supposed to
 * be as transparent as possible).
 *
 * The target thread inherits our MP count (added to its own) for the
 * duration of the preemption in order to preserve the atomicy of the
 * MP lock during the preemption.  Therefore, any preempting targets must be
 * careful in regards to MP assertions.  Note that the MP count may be
 * out of sync with the physical mp_lock, but we do not have to preserve
 * the original ownership of the lock if it was out of synch (that is, we
 * can leave it synchronized on return).
 */
void
lwkt_preempt(thread_t ntd, int critpri)
{
    struct globaldata *gd = mycpu;
    thread_t td = gd->gd_curthread;
#ifdef SMP
    int mpheld;
    int savecnt;
#endif

    /*
     * The caller has put us in a critical section.  We can only preempt
     * if the caller of the caller was not in a critical section (basically
     * a local interrupt), as determined by the 'critpri' parameter.   If
     * we are unable to preempt 
     *
     * YYY The target thread must be in a critical section (else it must
     * inherit our critical section?  I dunno yet).
     */
    KASSERT(ntd->td_pri >= TDPRI_CRIT, ("BADCRIT0 %d", ntd->td_pri));

    need_resched();
    if (!_lwkt_wantresched(ntd, td)) {
        ++preempt_miss;
        return;
    }
    if ((td->td_pri & ~TDPRI_MASK) > critpri) {
        ++preempt_miss;
        return;
    }
#ifdef SMP
    if (ntd->td_gd != gd) {
        ++preempt_miss;
        return;
    }
#endif
    if (td == ntd || ((td->td_flags | ntd->td_flags) & TDF_PREEMPT_LOCK)) {
        ++preempt_weird;
        return;
    }
    if (ntd->td_preempted) {
        ++preempt_hit;
        return;
    }
#ifdef SMP
    /*
     * note: an interrupt might have occured just as we were transitioning
     * to or from the MP lock.  In this case td_mpcount will be pre-disposed
     * (non-zero) but not actually synchronized with the actual state of the
     * lock.  We can use it to imply an MP lock requirement for the
     * preemption but we cannot use it to test whether we hold the MP lock
     * or not.
     */
    savecnt = td->td_mpcount;
    mpheld = MP_LOCK_HELD();
    ntd->td_mpcount += td->td_mpcount;
    if (mpheld == 0 && ntd->td_mpcount && !cpu_try_mplock()) {
        ntd->td_mpcount -= td->td_mpcount;
        ++preempt_miss;
        return;
    }
#endif

    ++preempt_hit;
    ntd->td_preempted = td;
    td->td_flags |= TDF_PREEMPT_LOCK;
    td->td_switch(ntd);
    KKASSERT(ntd->td_preempted && (td->td_flags & TDF_PREEMPT_DONE));
#ifdef SMP
    KKASSERT(savecnt == td->td_mpcount);
    mpheld = MP_LOCK_HELD();
    if (mpheld && td->td_mpcount == 0)
        cpu_rel_mplock();
    else if (mpheld == 0 && td->td_mpcount)
        panic("lwkt_preempt(): MP lock was not held through");
#endif
    ntd->td_preempted = NULL;
    td->td_flags &= ~(TDF_PREEMPT_LOCK|TDF_PREEMPT_DONE);
}

/*
 * Yield our thread while higher priority threads are pending.  This is
 * typically called when we leave a critical section but it can be safely
 * called while we are in a critical section.
 *
 * This function will not generally yield to equal priority threads but it
 * can occur as a side effect.  Note that lwkt_switch() is called from
 * inside the critical section to prevent its own crit_exit() from reentering
 * lwkt_yield_quick().
 *
 * gd_reqflags indicates that *something* changed, e.g. an interrupt or softint
 * came along but was blocked and made pending.
 *
 * (self contained on a per cpu basis)
 */
void
lwkt_yield_quick(void)
{
    globaldata_t gd = mycpu;
    thread_t td = gd->gd_curthread;

    /*
     * gd_reqflags is cleared in splz if the cpl is 0.  If we were to clear
     * it with a non-zero cpl then we might not wind up calling splz after
     * a task switch when the critical section is exited even though the
     * new task could accept the interrupt.
     *
     * XXX from crit_exit() only called after last crit section is released.
     * If called directly will run splz() even if in a critical section.
     *
     * td_nest_count prevent deep nesting via splz() or doreti().  Note that
     * except for this special case, we MUST call splz() here to handle any
     * pending ints, particularly after we switch, or we might accidently
     * halt the cpu with interrupts pending.
     */
    if (gd->gd_reqflags && td->td_nest_count < 2)
        splz();

    /*
     * YYY enabling will cause wakeup() to task-switch, which really
     * confused the old 4.x code.  This is a good way to simulate
     * preemption and MP without actually doing preemption or MP, because a
     * lot of code assumes that wakeup() does not block.
     */
    if (untimely_switch && td->td_nest_count == 0 &&
        gd->gd_intr_nesting_level == 0
    ) {
        crit_enter();
        /*
         * YYY temporary hacks until we disassociate the userland scheduler
         * from the LWKT scheduler.
         */
        if (td->td_flags & TDF_RUNQ) {
            lwkt_switch();              /* will not reenter yield function */
        } else {
            lwkt_schedule_self();       /* make sure we are scheduled */
            lwkt_switch();              /* will not reenter yield function */
            lwkt_deschedule_self();     /* make sure we are descheduled */
        }
        crit_exit_noyield(td);
    }
}

/*
 * This implements a normal yield which, unlike _quick, will yield to equal
 * priority threads as well.  Note that gd_reqflags tests will be handled by
 * the crit_exit() call in lwkt_switch().
 *
 * (self contained on a per cpu basis)
 */
void
lwkt_yield(void)
{
    lwkt_schedule_self();
    lwkt_switch();
}

/*
 * Schedule a thread to run.  As the current thread we can always safely
 * schedule ourselves, and a shortcut procedure is provided for that
 * function.
 *
 * (non-blocking, self contained on a per cpu basis)
 */
void
lwkt_schedule_self(void)
{
    thread_t td = curthread;

    crit_enter();
    KASSERT(td->td_wait == NULL, ("lwkt_schedule_self(): td_wait not NULL!"));
    _lwkt_enqueue(td);
#ifdef _KERNEL
    if (td->td_proc && td->td_proc->p_stat == SSLEEP)
        panic("SCHED SELF PANIC");
#endif
    crit_exit();
}

/*
 * Generic schedule.  Possibly schedule threads belonging to other cpus and
 * deal with threads that might be blocked on a wait queue.
 *
 * YYY this is one of the best places to implement load balancing code.
 * Load balancing can be accomplished by requesting other sorts of actions
 * for the thread in question.
 */
void
lwkt_schedule(thread_t td)
{
#ifdef  INVARIANTS
    if ((td->td_flags & TDF_PREEMPT_LOCK) == 0 && td->td_proc 
        && td->td_proc->p_stat == SSLEEP
    ) {
        printf("PANIC schedule curtd = %p (%d %d) target %p (%d %d)\n",
            curthread,
            curthread->td_proc ? curthread->td_proc->p_pid : -1,
            curthread->td_proc ? curthread->td_proc->p_stat : -1,
            td,
            td->td_proc ? curthread->td_proc->p_pid : -1,
            td->td_proc ? curthread->td_proc->p_stat : -1
        );
        panic("SCHED PANIC");
    }
#endif
    crit_enter();
    if (td == curthread) {
        _lwkt_enqueue(td);
    } else {
        lwkt_wait_t w;

        /*
         * If the thread is on a wait list we have to send our scheduling
         * request to the owner of the wait structure.  Otherwise we send
         * the scheduling request to the cpu owning the thread.  Races
         * are ok, the target will forward the message as necessary (the
         * message may chase the thread around before it finally gets
         * acted upon).
         *
         * (remember, wait structures use stable storage)
         */
        if ((w = td->td_wait) != NULL) {
            if (lwkt_trytoken(&w->wa_token)) {
                TAILQ_REMOVE(&w->wa_waitq, td, td_threadq);
                --w->wa_count;
                td->td_wait = NULL;
                if (smp_active == 0 || td->td_gd == mycpu) {
                    _lwkt_enqueue(td);
                    if (td->td_preemptable) {
                        td->td_preemptable(td, TDPRI_CRIT*2); /* YYY +token */
                    } else if (_lwkt_wantresched(td, curthread)) {
                        need_resched();
                    }
                } else {
                    lwkt_send_ipiq(td->td_gd->gd_cpuid, (ipifunc_t)lwkt_schedule, td);
                }
                lwkt_reltoken(&w->wa_token);
            } else {
                lwkt_send_ipiq(w->wa_token.t_cpu, (ipifunc_t)lwkt_schedule, td);
            }
        } else {
            /*
             * If the wait structure is NULL and we own the thread, there
             * is no race (since we are in a critical section).  If we
             * do not own the thread there might be a race but the
             * target cpu will deal with it.
             */
            if (smp_active == 0 || td->td_gd == mycpu) {
                _lwkt_enqueue(td);
                if (td->td_preemptable) {
                    td->td_preemptable(td, TDPRI_CRIT);
                } else if (_lwkt_wantresched(td, curthread)) {
                    need_resched();
                }
            } else {
                lwkt_send_ipiq(td->td_gd->gd_cpuid, (ipifunc_t)lwkt_schedule, td);
            }
        }
    }
    crit_exit();
}

/*
 * Managed acquisition.  This code assumes that the MP lock is held for
 * the tdallq operation and that the thread has been descheduled from its
 * original cpu.  We also have to wait for the thread to be entirely switched
 * out on its original cpu (this is usually fast enough that we never loop)
 * since the LWKT system does not have to hold the MP lock while switching
 * and the target may have released it before switching.
 */
void
lwkt_acquire(thread_t td)
{
    struct globaldata *gd;

    gd = td->td_gd;
    KKASSERT((td->td_flags & TDF_RUNQ) == 0);
    while (td->td_flags & TDF_RUNNING)  /* XXX spin */
        ;
    if (gd != mycpu) {
        crit_enter();
        TAILQ_REMOVE(&gd->gd_tdallq, td, td_allq);      /* protected by BGL */
        gd = mycpu;
        td->td_gd = gd;
        TAILQ_INSERT_TAIL(&gd->gd_tdallq, td, td_allq); /* protected by BGL */
        crit_exit();
    }
}

/*
 * Deschedule a thread.
 *
 * (non-blocking, self contained on a per cpu basis)
 */
void
lwkt_deschedule_self(void)
{
    thread_t td = curthread;

    crit_enter();
    KASSERT(td->td_wait == NULL, ("lwkt_schedule_self(): td_wait not NULL!"));
    _lwkt_dequeue(td);
    crit_exit();
}

/*
 * Generic deschedule.  Descheduling threads other then your own should be
 * done only in carefully controlled circumstances.  Descheduling is 
 * asynchronous.  
 *
 * This function may block if the cpu has run out of messages.
 */
void
lwkt_deschedule(thread_t td)
{
    crit_enter();
    if (td == curthread) {
        _lwkt_dequeue(td);
    } else {
        if (td->td_gd == mycpu) {
            _lwkt_dequeue(td);
        } else {
            lwkt_send_ipiq(td->td_gd->gd_cpuid, (ipifunc_t)lwkt_deschedule, td);
        }
    }
    crit_exit();
}

/*
 * Set the target thread's priority.  This routine does not automatically
 * switch to a higher priority thread, LWKT threads are not designed for
 * continuous priority changes.  Yield if you want to switch.
 *
 * We have to retain the critical section count which uses the high bits
 * of the td_pri field.  The specified priority may also indicate zero or
 * more critical sections by adding TDPRI_CRIT*N.
 */
void
lwkt_setpri(thread_t td, int pri)
{
    KKASSERT(pri >= 0);
    KKASSERT(td->td_gd == mycpu);
    crit_enter();
    if (td->td_flags & TDF_RUNQ) {
        _lwkt_dequeue(td);
        td->td_pri = (td->td_pri & ~TDPRI_MASK) + pri;
        _lwkt_enqueue(td);
    } else {
        td->td_pri = (td->td_pri & ~TDPRI_MASK) + pri;
    }
    crit_exit();
}

void
lwkt_setpri_self(int pri)
{
    thread_t td = curthread;

    KKASSERT(pri >= 0 && pri <= TDPRI_MAX);
    crit_enter();
    if (td->td_flags & TDF_RUNQ) {
        _lwkt_dequeue(td);
        td->td_pri = (td->td_pri & ~TDPRI_MASK) + pri;
        _lwkt_enqueue(td);
    } else {
        td->td_pri = (td->td_pri & ~TDPRI_MASK) + pri;
    }
    crit_exit();
}

struct proc *
lwkt_preempted_proc(void)
{
    thread_t td = curthread;
    while (td->td_preempted)
        td = td->td_preempted;
    return(td->td_proc);
}

typedef struct lwkt_gettoken_req {
    lwkt_token_t tok;
    int cpu;
} lwkt_gettoken_req;

#if 0

/*
 * This function deschedules the current thread and blocks on the specified
 * wait queue.  We obtain ownership of the wait queue in order to block
 * on it.  A generation number is used to interlock the wait queue in case
 * it gets signalled while we are blocked waiting on the token.
 *
 * Note: alternatively we could dequeue our thread and then message the
 * target cpu owning the wait queue.  YYY implement as sysctl.
 *
 * Note: wait queue signals normally ping-pong the cpu as an optimization.
 */

void
lwkt_block(lwkt_wait_t w, const char *wmesg, int *gen)
{
    thread_t td = curthread;

    lwkt_gettoken(&w->wa_token);
    if (w->wa_gen == *gen) {
        _lwkt_dequeue(td);
        TAILQ_INSERT_TAIL(&w->wa_waitq, td, td_threadq);
        ++w->wa_count;
        td->td_wait = w;
        td->td_wmesg = wmesg;
again:
        lwkt_switch();
        lwkt_regettoken(&w->wa_token);
        if (td->td_wmesg != NULL) {
            _lwkt_dequeue(td);
            goto again;
        }
    }
    /* token might be lost, doesn't matter for gen update */
    *gen = w->wa_gen;
    lwkt_reltoken(&w->wa_token);
}

/*
 * Signal a wait queue.  We gain ownership of the wait queue in order to
 * signal it.  Once a thread is removed from the wait queue we have to
 * deal with the cpu owning the thread.
 *
 * Note: alternatively we could message the target cpu owning the wait
 * queue.  YYY implement as sysctl.
 */
void
lwkt_signal(lwkt_wait_t w, int count)
{
    thread_t td;
    int count;

    lwkt_gettoken(&w->wa_token);
    ++w->wa_gen;
    if (count < 0)
        count = w->wa_count;
    while ((td = TAILQ_FIRST(&w->wa_waitq)) != NULL && count) {
        --count;
        --w->wa_count;
        TAILQ_REMOVE(&w->wa_waitq, td, td_threadq);
        td->td_wait = NULL;
        td->td_wmesg = NULL;
        if (td->td_gd == mycpu) {
            _lwkt_enqueue(td);
        } else {
            lwkt_send_ipiq(td->td_gd->gd_cpuid, (ipifunc_t)lwkt_schedule, td);
        }
        lwkt_regettoken(&w->wa_token);
    }
    lwkt_reltoken(&w->wa_token);
}

#endif

/*
 * Acquire ownership of a token
 *
 * Acquire ownership of a token.  The token may have spl and/or critical
 * section side effects, depending on its purpose.  These side effects
 * guarentee that you will maintain ownership of the token as long as you
 * do not block.  If you block you may lose access to the token (but you
 * must still release it even if you lose your access to it).
 *
 * YYY for now we use a critical section to prevent IPIs from taking away
 * a token, but do we really only need to disable IPIs ?
 *
 * YYY certain tokens could be made to act like mutexes when performance
 * would be better (e.g. t_cpu == -1).  This is not yet implemented.
 *
 * YYY the tokens replace 4.x's simplelocks for the most part, but this
 * means that 4.x does not expect a switch so for now we cannot switch
 * when waiting for an IPI to be returned.  
 *
 * YYY If the token is owned by another cpu we may have to send an IPI to
 * it and then block.   The IPI causes the token to be given away to the
 * requesting cpu, unless it has already changed hands.  Since only the
 * current cpu can give away a token it owns we do not need a memory barrier.
 * This needs serious optimization.
 */

#ifdef SMP

static
void
lwkt_gettoken_remote(void *arg)
{
    lwkt_gettoken_req *req = arg;
    if (req->tok->t_cpu == mycpu->gd_cpuid) {
#ifdef INVARIANTS
        if (token_debug)
            printf("GT(%d,%d) ", req->tok->t_cpu, req->cpu);
#endif
        req->tok->t_cpu = req->cpu;
        req->tok->t_reqcpu = req->cpu;  /* YYY leave owned by target cpu */
        /* else set reqcpu to point to current cpu for release */
    }
}

#endif

int
lwkt_gettoken(lwkt_token_t tok)
{
    /*
     * Prevent preemption so the token can't be taken away from us once
     * we gain ownership of it.  Use a synchronous request which might
     * block.  The request will be forwarded as necessary playing catchup
     * to the token.
     */

    crit_enter();
#ifdef INVARIANTS
    if (curthread->td_pri > 1800) {
        printf("lwkt_gettoken: %p called from %p: crit sect nesting warning\n",
            tok, ((int **)&tok)[-1]);
    }
    if (curthread->td_pri > 2000) {
        curthread->td_pri = 1000;
        panic("too HIGH!");
    }
#endif
#ifdef SMP
    while (tok->t_cpu != mycpu->gd_cpuid) {
        struct lwkt_gettoken_req req;
        int seq;
        int dcpu;

        req.cpu = mycpu->gd_cpuid;
        req.tok = tok;
        dcpu = (volatile int)tok->t_cpu;
        KKASSERT(dcpu >= 0 && dcpu < ncpus);
#ifdef INVARIANTS
        if (token_debug)
            printf("REQT%d ", dcpu);
#endif
        seq = lwkt_send_ipiq(dcpu, lwkt_gettoken_remote, &req);
        lwkt_wait_ipiq(dcpu, seq);
#ifdef INVARIANTS
        if (token_debug)
            printf("REQR%d ", tok->t_cpu);
#endif
    }
#endif
    /*
     * leave us in a critical section on return.  This will be undone
     * by lwkt_reltoken().  Bump the generation number.
     */
    return(++tok->t_gen);
}

/*
 * Attempt to acquire ownership of a token.  Returns 1 on success, 0 on
 * failure.
 */
int
lwkt_trytoken(lwkt_token_t tok)
{
    crit_enter();
#ifdef SMP
    if (tok->t_cpu != mycpu->gd_cpuid) {
        crit_exit();
        return(0);
    } 
#endif
    /* leave us in the critical section */
    ++tok->t_gen;
    return(1);
}

/*
 * Release your ownership of a token.  Releases must occur in reverse
 * order to aquisitions, eventually so priorities can be unwound properly
 * like SPLs.  At the moment the actual implemention doesn't care.
 *
 * We can safely hand a token that we own to another cpu without notifying
 * it, but once we do we can't get it back without requesting it (unless
 * the other cpu hands it back to us before we check).
 *
 * We might have lost the token, so check that.
 *
 * Return the token's generation number.  The number is useful to callers
 * who may want to know if the token was stolen during potential blockages.
 */
int
lwkt_reltoken(lwkt_token_t tok)
{
    int gen;

    if (tok->t_cpu == mycpu->gd_cpuid) {
        tok->t_cpu = tok->t_reqcpu;
    }
    gen = tok->t_gen;
    crit_exit();
    return(gen);
}

/*
 * Reacquire a token that might have been lost.  0 is returned if the 
 * generation has not changed (nobody stole the token from us), -1 is 
 * returned otherwise.  The token is reacquired regardless but the
 * generation number is not bumped further if we already own the token.
 *
 * For efficiency we inline the best-case situation for lwkt_regettoken()
 * (i.e .we still own the token).
 */
int
lwkt_gentoken(lwkt_token_t tok, int *gen)
{
    if (tok->t_cpu == mycpu->gd_cpuid && tok->t_gen == *gen)
        return(0);
    *gen = lwkt_regettoken(tok);
    return(-1);
}

/*
 * Re-acquire a token that might have been lost.   The generation number
 * is bumped and returned regardless of whether the token had been lost
 * or not (because we only have cpu granularity we have to bump the token
 * either way).
 */
int
lwkt_regettoken(lwkt_token_t tok)
{
    /* assert we are in a critical section */
    if (tok->t_cpu != mycpu->gd_cpuid) {
#ifdef SMP
        while (tok->t_cpu != mycpu->gd_cpuid) {
            struct lwkt_gettoken_req req;
            int seq;
            int dcpu;

            req.cpu = mycpu->gd_cpuid;
            req.tok = tok;
            dcpu = (volatile int)tok->t_cpu;
            KKASSERT(dcpu >= 0 && dcpu < ncpus);
#ifdef INVARIANTS
            if (token_debug)
                printf("REQT%d ", dcpu);
#endif
            seq = lwkt_send_ipiq(dcpu, lwkt_gettoken_remote, &req);
            lwkt_wait_ipiq(dcpu, seq);
#ifdef INVARIATNS
            if (token_debug)
                printf("REQR%d ", tok->t_cpu);
#endif
        }
#endif
    }
    ++tok->t_gen;
    return(tok->t_gen);
}

void
lwkt_inittoken(lwkt_token_t tok)
{
    /*
     * Zero structure and set cpu owner and reqcpu to cpu 0.
     */
    bzero(tok, sizeof(*tok));
}

/*
 * Create a kernel process/thread/whatever.  It shares it's address space
 * with proc0 - ie: kernel only.
 *
 * NOTE!  By default new threads are created with the MP lock held.  A 
 * thread which does not require the MP lock should release it by calling
 * rel_mplock() at the start of the new thread.
 */
int
lwkt_create(void (*func)(void *), void *arg,
    struct thread **tdp, thread_t template, int tdflags, int cpu,
    const char *fmt, ...)
{
    thread_t td;
    __va_list ap;

    td = lwkt_alloc_thread(template, cpu);
    if (tdp)
        *tdp = td;
    cpu_set_thread_handler(td, kthread_exit, func, arg);
    td->td_flags |= TDF_VERBOSE | tdflags;
#ifdef SMP
    td->td_mpcount = 1;
#endif

    /*
     * Set up arg0 for 'ps' etc
     */
    __va_start(ap, fmt);
    vsnprintf(td->td_comm, sizeof(td->td_comm), fmt, ap);
    __va_end(ap);

    /*
     * Schedule the thread to run
     */
    if ((td->td_flags & TDF_STOPREQ) == 0)
        lwkt_schedule(td);
    else
        td->td_flags &= ~TDF_STOPREQ;
    return 0;
}

/*
 * Destroy an LWKT thread.   Warning!  This function is not called when
 * a process exits, cpu_proc_exit() directly calls cpu_thread_exit() and
 * uses a different reaping mechanism.
 */
void
lwkt_exit(void)
{
    thread_t td = curthread;

    if (td->td_flags & TDF_VERBOSE)
        printf("kthread %p %s has exited\n", td, td->td_comm);
    crit_enter();
    lwkt_deschedule_self();
    ++mycpu->gd_tdfreecount;
    TAILQ_INSERT_TAIL(&mycpu->gd_tdfreeq, td, td_threadq);
    cpu_thread_exit();
}

#ifdef _KERNEL
/*
 * Create a kernel process/thread/whatever.  It shares it's address space
 * with proc0 - ie: kernel only.  5.x compatible.
 *
 * NOTE!  By default kthreads are created with the MP lock held.  A
 * thread which does not require the MP lock should release it by calling
 * rel_mplock() at the start of the new thread.
 */
int
kthread_create(void (*func)(void *), void *arg,
    struct thread **tdp, const char *fmt, ...)
{
    thread_t td;
    __va_list ap;

    td = lwkt_alloc_thread(NULL, -1);
    if (tdp)
        *tdp = td;
    cpu_set_thread_handler(td, kthread_exit, func, arg);
    td->td_flags |= TDF_VERBOSE;
#ifdef SMP
    td->td_mpcount = 1;
#endif

    /*
     * Set up arg0 for 'ps' etc
     */
    __va_start(ap, fmt);
    vsnprintf(td->td_comm, sizeof(td->td_comm), fmt, ap);
    __va_end(ap);

    /*
     * Schedule the thread to run
     */
    lwkt_schedule(td);
    return 0;
}

#endif

void
crit_panic(void)
{
    thread_t td = curthread;
    int lpri = td->td_pri;

    td->td_pri = 0;
    panic("td_pri is/would-go negative! %p %d", td, lpri);
}

/*
 * Destroy an LWKT thread.   Warning!  This function is not called when
 * a process exits, cpu_proc_exit() directly calls cpu_thread_exit() and
 * uses a different reaping mechanism.
 *
 * XXX duplicates lwkt_exit()
 */
void
kthread_exit(void)
{
    lwkt_exit();
}

#ifdef SMP

/*
 * Send a function execution request to another cpu.  The request is queued
 * on the cpu<->cpu ipiq matrix.  Each cpu owns a unique ipiq FIFO for every
 * possible target cpu.  The FIFO can be written.
 *
 * YYY If the FIFO fills up we have to enable interrupts and process the
 * IPIQ while waiting for it to empty or we may deadlock with another cpu.
 * Create a CPU_*() function to do this!
 *
 * We can safely bump gd_intr_nesting_level because our crit_exit() at the
 * end will take care of any pending interrupts.
 *
 * Must be called from a critical section.
 */
int
lwkt_send_ipiq(int dcpu, ipifunc_t func, void *arg)
{
    lwkt_ipiq_t ip;
    int windex;
    struct globaldata *gd = mycpu;

    if (dcpu == gd->gd_cpuid) {
        func(arg);
        return(0);
    } 
    crit_enter();
    ++gd->gd_intr_nesting_level;
#ifdef INVARIANTS
    if (gd->gd_intr_nesting_level > 20)
        panic("lwkt_send_ipiq: TOO HEAVILY NESTED!");
#endif
    KKASSERT(curthread->td_pri >= TDPRI_CRIT);
    KKASSERT(dcpu >= 0 && dcpu < ncpus);
    ++ipiq_count;
    ip = &gd->gd_ipiq[dcpu];

    /*
     * We always drain before the FIFO becomes full so it should never
     * become full.  We need to leave enough entries to deal with 
     * reentrancy.
     */
    KKASSERT(ip->ip_windex - ip->ip_rindex != MAXCPUFIFO);
    windex = ip->ip_windex & MAXCPUFIFO_MASK;
    ip->ip_func[windex] = func;
    ip->ip_arg[windex] = arg;
    /* YYY memory barrier */
    ++ip->ip_windex;
    if (ip->ip_windex - ip->ip_rindex > MAXCPUFIFO / 2) {
        unsigned int eflags = read_eflags();
        cpu_enable_intr();
        ++ipiq_fifofull;
        while (ip->ip_windex - ip->ip_rindex > MAXCPUFIFO / 4) {
            KKASSERT(ip->ip_windex - ip->ip_rindex != MAXCPUFIFO - 1);
            lwkt_process_ipiq();
        }
        write_eflags(eflags);
    }
    --gd->gd_intr_nesting_level;
    cpu_send_ipiq(dcpu);        /* issues memory barrier if appropriate */
    crit_exit();
    return(ip->ip_windex);
}

/*
 * Send a message to several target cpus.  Typically used for scheduling.
 * The message will not be sent to stopped cpus.
 */
void
lwkt_send_ipiq_mask(u_int32_t mask, ipifunc_t func, void *arg)
{
    int cpuid;

    mask &= ~stopped_cpus;
    while (mask) {
            cpuid = bsfl(mask);
            lwkt_send_ipiq(cpuid, func, arg);
            mask &= ~(1 << cpuid);
    }
}

/*
 * Wait for the remote cpu to finish processing a function.
 *
 * YYY we have to enable interrupts and process the IPIQ while waiting
 * for it to empty or we may deadlock with another cpu.  Create a CPU_*()
 * function to do this!  YYY we really should 'block' here.
 *
 * Must be called from a critical section.  Thsi routine may be called
 * from an interrupt (for example, if an interrupt wakes a foreign thread
 * up).
 */
void
lwkt_wait_ipiq(int dcpu, int seq)
{
    lwkt_ipiq_t ip;
    int maxc = 100000000;

    if (dcpu != mycpu->gd_cpuid) {
        KKASSERT(dcpu >= 0 && dcpu < ncpus);
        ip = &mycpu->gd_ipiq[dcpu];
        if ((int)(ip->ip_xindex - seq) < 0) {
            unsigned int eflags = read_eflags();
            cpu_enable_intr();
            while ((int)(ip->ip_xindex - seq) < 0) {
                lwkt_process_ipiq();
                if (--maxc == 0)
                        printf("LWKT_WAIT_IPIQ WARNING! %d wait %d (%d)\n", mycpu->gd_cpuid, dcpu, ip->ip_xindex - seq);
                if (maxc < -1000000)
                        panic("LWKT_WAIT_IPIQ");
            }
            write_eflags(eflags);
        }
    }
}

/*
 * Called from IPI interrupt (like a fast interrupt), which has placed
 * us in a critical section.  The MP lock may or may not be held.
 * May also be called from doreti or splz, or be reentrantly called
 * indirectly through the ip_func[] we run.
 */
void
lwkt_process_ipiq(void)
{
    int n;
    int cpuid = mycpu->gd_cpuid;

    for (n = 0; n < ncpus; ++n) {
        lwkt_ipiq_t ip;
        int ri;

        if (n == cpuid)
            continue;
        ip = globaldata_find(n)->gd_ipiq;
        if (ip == NULL)
            continue;
        ip = &ip[cpuid];

        /*
         * Note: xindex is only updated after we are sure the function has
         * finished execution.  Beware lwkt_process_ipiq() reentrancy!  The
         * function may send an IPI which may block/drain.
         */
        while (ip->ip_rindex != ip->ip_windex) {
            ri = ip->ip_rindex & MAXCPUFIFO_MASK;
            ++ip->ip_rindex;
            ip->ip_func[ri](ip->ip_arg[ri]);
            /* YYY memory barrier */
            ip->ip_xindex = ip->ip_rindex;
        }
    }
}

#else

int
lwkt_send_ipiq(int dcpu, ipifunc_t func, void *arg)
{
    panic("lwkt_send_ipiq: UP box! (%d,%p,%p)", dcpu, func, arg);
    return(0); /* NOT REACHED */
}

void
lwkt_wait_ipiq(int dcpu, int seq)
{
    panic("lwkt_wait_ipiq: UP box! (%d,%d)", dcpu, seq);
}

#endif