route: addroute may change rt->rt_ifa; keep the local copy in sync.

[dragonfly.git] / usr.bin / top / m_dragonfly.c

/*
 * top - a top users display for Unix
 *
 * SYNOPSIS:  For DragonFly 2.x and later
 *
 * DESCRIPTION:
 * Originally written for BSD4.4 system by Christos Zoulas.
 * Ported to FreeBSD 2.x by Steven Wallace && Wolfram Schneider
 * Order support hacked in from top-3.5beta6/machine/m_aix41.c
 *   by Monte Mitzelfelt (for latest top see http://www.groupsys.com/topinfo/)
 *
 * This is the machine-dependent module for DragonFly 2.5.1
 * Should work for:
 *      DragonFly 2.x and above
 *
 * LIBS: -lkvm
 *
 * AUTHOR: Jan Lentfer <Jan.Lentfer@web.de>
 * This module has been put together from different sources and is based on the
 * work of many other people, e.g. Matthew Dillon, Simon Schubert, Jordan Gordeev.
 *
 * $FreeBSD: src/usr.bin/top/machine.c,v 1.29.2.2 2001/07/31 20:27:05 tmm Exp $
 */

#include <sys/user.h>
#include <sys/types.h>
#include <sys/time.h>
#include <sys/signal.h>
#include <sys/param.h>

#include "os.h"
#include <err.h>
#include <kvm.h>
#include <stdio.h>
#include <unistd.h>
#include <math.h>
#include <pwd.h>
#include <sys/errno.h>
#include <sys/sysctl.h>
#include <sys/file.h>
#include <sys/vmmeter.h>
#include <sys/resource.h>
#include <sys/rtprio.h>

/* Swap */
#include <stdlib.h>
#include <string.h>
#include <sys/conf.h>

#include <osreldate.h>          /* for changes in kernel structures */

#include <sys/kinfo.h>
#include <kinfo.h>
#include "top.h"
#include "display.h"
#include "machine.h"
#include "screen.h"
#include "utils.h"

int swapmode(int *retavail, int *retfree);
static int namelength;
static int cmdlength;
static int show_fullcmd;

int n_cpus = 0;

/* get_process_info passes back a handle.  This is what it looks like: */

struct handle {
        struct kinfo_proc **next_proc;  /* points to next valid proc pointer */
        int remaining;          /* number of pointers remaining */
        int show_threads;
};

/* declarations for load_avg */
#include "loadavg.h"

#define PP(pp, field) ((pp)->kp_ ## field)
#define LP(pp, field) ((pp)->kp_lwp.kl_ ## field)
#define VP(pp, field) ((pp)->kp_vm_ ## field)

/* what we consider to be process size: */
#define PROCSIZE(pp) (VP((pp), map_size) / 1024)

/*
 * These definitions control the format of the per-process area
 */

static char smp_header[] =
"   PID %-*.*s NICE  SIZE    RES    STATE   C   TIME   CTIME    CPU COMMAND";

#define smp_Proc_format \
        "%6d %-*.*s %3d%7s %6s %8.8s %3d %6s %7s %5.2f%% %.*s"

/* process state names for the "STATE" column of the display */
/*
 * the extra nulls in the string "run" are for adding a slash and the
 * processor number when needed
 */

const char *state_abbrev[] = {
        "", "RUN\0\0\0", "STOP", "SLEEP",
};


static kvm_t *kd;

/* values that we stash away in _init and use in later routines */

static long lastpid;

/* these are for calculating cpu state percentages */

static struct kinfo_cputime *cp_time, *cp_old;

/* these are for detailing the process states */

#define MAXPSTATES      6

int process_states[MAXPSTATES];

char *procstatenames[] = {
        " running, ", " idle, ", " active, ", " stopped, ", " zombie, ",
        NULL
};

/* these are for detailing the cpu states */
#define CPU_STATES 5
int *cpu_states;
int* cpu_averages;
char *cpustatenames[CPU_STATES + 1] = {
        "user", "nice", "system", "interrupt", "idle", NULL
};

/* these are for detailing the memory statistics */

long memory_stats[7];
char *memorynames[] = {
        "K Active, ", "K Inact, ", "K Wired, ", "K Cache, ", "K Buf, ", "K Free",
        NULL
};

long swap_stats[7];
char *swapnames[] = {
        /* 0           1            2           3            4       5 */
        "K Total, ", "K Used, ", "K Free, ", "% Inuse, ", "K In, ", "K Out",
        NULL
};


/* these are for keeping track of the proc array */

static int nproc;
static int onproc = -1;
static int pref_len;
static struct kinfo_proc *pbase;
static struct kinfo_proc **pref;

static uint64_t prev_pbase_time;        /* unit: us */
static struct kinfo_proc *prev_pbase;
static int prev_pbase_alloc;
static int prev_nproc;
static int fscale;

/* these are for getting the memory statistics */

static int pageshift;           /* log base 2 of the pagesize */

/* define pagetok in terms of pageshift */

#define pagetok(size) ((size) << pageshift)

/* sorting orders. first is default */
char *ordernames[] = {
  "cpu", "size", "res", "time", "pri", "thr", "pid", "ctime",  "pres", NULL
};

/* compare routines */
int proc_compare (struct kinfo_proc **, struct kinfo_proc **);
int compare_size (struct kinfo_proc **, struct kinfo_proc **);
int compare_res (struct kinfo_proc **, struct kinfo_proc **);
int compare_time (struct kinfo_proc **, struct kinfo_proc **);
int compare_ctime (struct kinfo_proc **, struct kinfo_proc **);
int compare_prio(struct kinfo_proc **, struct kinfo_proc **);
int compare_thr (struct kinfo_proc **, struct kinfo_proc **);
int compare_pid (struct kinfo_proc **, struct kinfo_proc **);
int compare_pres(struct kinfo_proc **, struct kinfo_proc **);

int (*proc_compares[]) (struct kinfo_proc **,struct kinfo_proc **) = {
        proc_compare,
        compare_size,
        compare_res,
        compare_time,
        compare_prio,
        compare_thr,
        compare_pid,
        compare_ctime,
        compare_pres,
        NULL
};

static void
cputime_percentages(int out[CPU_STATES], struct kinfo_cputime *new,
    struct kinfo_cputime *old)
{
        struct kinfo_cputime diffs;
        uint64_t total_change, half_total;

        /* initialization */
        total_change = 0;

        diffs.cp_user = new->cp_user - old->cp_user;
        diffs.cp_nice = new->cp_nice - old->cp_nice;
        diffs.cp_sys = new->cp_sys - old->cp_sys;
        diffs.cp_intr = new->cp_intr - old->cp_intr;
        diffs.cp_idle = new->cp_idle - old->cp_idle;
        total_change = diffs.cp_user + diffs.cp_nice + diffs.cp_sys +
            diffs.cp_intr + diffs.cp_idle;
        old->cp_user = new->cp_user;
        old->cp_nice = new->cp_nice;
        old->cp_sys = new->cp_sys;
        old->cp_intr = new->cp_intr;
        old->cp_idle = new->cp_idle;

        /* avoid divide by zero potential */
        if (total_change == 0)
                total_change = 1;

        /* calculate percentages based on overall change, rounding up */
        half_total = total_change >> 1;

        out[0] = ((diffs.cp_user * 1000LL + half_total) / total_change);
        out[1] = ((diffs.cp_nice * 1000LL + half_total) / total_change);
        out[2] = ((diffs.cp_sys * 1000LL + half_total) / total_change);
        out[3] = ((diffs.cp_intr * 1000LL + half_total) / total_change);
        out[4] = ((diffs.cp_idle * 1000LL + half_total) / total_change);
}

int
machine_init(struct statics *statics)
{
        int pagesize;
        size_t modelen, prmlen;
        struct passwd *pw;
        struct timeval boottime;

        if (n_cpus < 1) {
                if (kinfo_get_cpus(&n_cpus))
                        err(1, "kinfo_get_cpus failed");
        }
        /* get boot time */
        modelen = sizeof(boottime);
        if (sysctlbyname("kern.boottime", &boottime, &modelen, NULL, 0) == -1) {
                /* we have no boottime to report */
                boottime.tv_sec = -1;
        }

        prmlen = sizeof(fscale);
        if (sysctlbyname("kern.fscale", &fscale, &prmlen, NULL, 0) == -1)
                err(1, "sysctl kern.fscale failed");

        while ((pw = getpwent()) != NULL) {
                if ((int)strlen(pw->pw_name) > namelength)
                        namelength = strlen(pw->pw_name);
        }
        if (namelength < 8)
                namelength = 8;
        if (namelength > 13)
                namelength = 13;

        if ((kd = kvm_open(NULL, NULL, NULL, O_RDONLY, NULL)) == NULL)
                return -1;

        pbase = NULL;
        pref = NULL;
        nproc = 0;
        onproc = -1;
        prev_pbase = NULL;
        prev_pbase_alloc = 0;
        prev_pbase_time = 0;
        prev_nproc = 0;
        /*
         * get the page size with "getpagesize" and calculate pageshift from
         * it
         */
        pagesize = getpagesize();
        pageshift = 0;
        while (pagesize > 1) {
                pageshift++;
                pagesize >>= 1;
        }

        /* we only need the amount of log(2)1024 for our conversion */
        pageshift -= LOG1024;

        /* fill in the statics information */
        statics->procstate_names = procstatenames;
        statics->cpustate_names = cpustatenames;
        statics->memory_names = memorynames;
        statics->boottime = boottime.tv_sec;
        statics->swap_names = swapnames;
        statics->order_names = ordernames;
        /* we need kvm descriptor in order to show full commands */
        statics->flags.fullcmds = kd != NULL;
        statics->flags.threads = 1;

        /* all done! */
        return (0);
}

char *
format_header(char *uname_field)
{
        static char Header[128];

        snprintf(Header, sizeof(Header), smp_header,
            namelength, namelength, uname_field);

        if (screen_width <= 79)
                cmdlength = 80;
        else
                cmdlength = screen_width;

        cmdlength = cmdlength - strlen(Header) + 6;

        return Header;
}

static int swappgsin = -1;
static int swappgsout = -1;
extern struct timeval timeout;

void
get_system_info(struct system_info *si)
{
        size_t len;
        int cpu;

        if (cpu_states == NULL) {
                cpu_states = malloc(sizeof(*cpu_states) * CPU_STATES * n_cpus);
                if (cpu_states == NULL)
                        err(1, "malloc");
                bzero(cpu_states, sizeof(*cpu_states) * CPU_STATES * n_cpus);
        }
        if (cp_time == NULL) {
                cp_time = malloc(2 * n_cpus * sizeof(cp_time[0]));
                if (cp_time == NULL)
                        err(1, "cp_time");
                cp_old = cp_time + n_cpus;
                len = n_cpus * sizeof(cp_old[0]);
                bzero(cp_time, len);
                if (sysctlbyname("kern.cputime", cp_old, &len, NULL, 0))
                        err(1, "kern.cputime");
        }
        len = n_cpus * sizeof(cp_time[0]);
        bzero(cp_time, len);
        if (sysctlbyname("kern.cputime", cp_time, &len, NULL, 0))
                err(1, "kern.cputime");

        getloadavg(si->load_avg, 3);

        lastpid = 0;

        /* convert cp_time counts to percentages */
        int combine_cpus = (enable_ncpus == 0 && n_cpus > 1);
        for (cpu = 0; cpu < n_cpus; ++cpu) {
                cputime_percentages(cpu_states + cpu * CPU_STATES,
                    &cp_time[cpu], &cp_old[cpu]);
        }
        if (combine_cpus) {
                if (cpu_averages == NULL) {
                        cpu_averages = malloc(sizeof(*cpu_averages) * CPU_STATES);
                        if (cpu_averages == NULL)
                                err(1, "cpu_averages");
                }
                bzero(cpu_averages, sizeof(*cpu_averages) * CPU_STATES);
                for (cpu = 0; cpu < n_cpus; ++cpu) {
                        int j = 0;
                        cpu_averages[0] += *(cpu_states + ((cpu * CPU_STATES) + j++) );
                        cpu_averages[1] += *(cpu_states + ((cpu * CPU_STATES) + j++) );
                        cpu_averages[2] += *(cpu_states + ((cpu * CPU_STATES) + j++) );
                        cpu_averages[3] += *(cpu_states + ((cpu * CPU_STATES) + j++) );
                        cpu_averages[4] += *(cpu_states + ((cpu * CPU_STATES) + j++) );
                }
                for (int i = 0; i < CPU_STATES; ++i)
                        cpu_averages[i] /= n_cpus;
        }

        /* sum memory & swap statistics */
        {
                struct vmmeter vmm;
                struct vmstats vms;
                size_t vms_size = sizeof(vms);
                size_t vmm_size = sizeof(vmm);
                static unsigned int swap_delay = 0;
                static int swapavail = 0;
                static int swapfree = 0;
                static long bufspace = 0;

                if (sysctlbyname("vm.vmstats", &vms, &vms_size, NULL, 0))
                        err(1, "sysctlbyname: vm.vmstats");

                if (sysctlbyname("vm.vmmeter", &vmm, &vmm_size, NULL, 0))
                        err(1, "sysctlbyname: vm.vmmeter");

                if (kinfo_get_vfs_bufspace(&bufspace))
                        err(1, "kinfo_get_vfs_bufspace");

                /* convert memory stats to Kbytes */
                memory_stats[0] = pagetok(vms.v_active_count);
                memory_stats[1] = pagetok(vms.v_inactive_count);
                memory_stats[2] = pagetok(vms.v_wire_count);
                memory_stats[3] = pagetok(vms.v_cache_count);
                memory_stats[4] = bufspace / 1024;
                memory_stats[5] = pagetok(vms.v_free_count);
                memory_stats[6] = -1;

                /* first interval */
                if (swappgsin < 0) {
                        swap_stats[4] = 0;
                        swap_stats[5] = 0;
                }
                /* compute differences between old and new swap statistic */
                else {
                        swap_stats[4] = pagetok(((vmm.v_swappgsin - swappgsin)));
                        swap_stats[5] = pagetok(((vmm.v_swappgsout - swappgsout)));
                }

                swappgsin = vmm.v_swappgsin;
                swappgsout = vmm.v_swappgsout;

                /* call CPU heavy swapmode() only for changes */
                if (swap_stats[4] > 0 || swap_stats[5] > 0 || swap_delay == 0) {
                        swap_stats[3] = swapmode(&swapavail, &swapfree);
                        swap_stats[0] = swapavail;
                        swap_stats[1] = swapavail - swapfree;
                        swap_stats[2] = swapfree;
                }
                swap_delay = 1;
                swap_stats[6] = -1;
        }

        /* set arrays and strings */
        si->cpustates = combine_cpus == 1 ?
            cpu_averages : cpu_states;
        si->memory = memory_stats;
        si->swap = swap_stats;


        if (lastpid > 0) {
                si->last_pid = lastpid;
        } else {
                si->last_pid = -1;
        }
}


static struct handle handle;

static void
fixup_pctcpu(struct kinfo_proc *fixit, uint64_t d)
{
        struct kinfo_proc *pp;
        uint64_t ticks;
        int i;

        if (prev_nproc == 0 || d == 0)
                return;

        if (LP(fixit, pid) == -1) {
                /* Skip kernel "idle" threads */
                if (PP(fixit, stat) == SIDL)
                        return;
                for (pp = prev_pbase, i = 0; i < prev_nproc; pp++, i++) {
                        if (LP(pp, pid) == -1 &&
                            PP(pp, ktaddr) == PP(fixit, ktaddr))
                                break;
                }
        } else {
                for (pp = prev_pbase, i = 0; i < prev_nproc; pp++, i++) {
                        if (LP(pp, pid) == LP(fixit, pid) &&
                            LP(pp, tid) == LP(fixit, tid)) {
                                if (PP(pp, paddr) != PP(fixit, paddr)) {
                                        /* pid/tid are reused */
                                        pp = NULL;
                                }
                                break;
                        }
                }
        }
        if (i == prev_nproc || pp == NULL)
                return;

        ticks = LP(fixit, iticks) - LP(pp, iticks);
        ticks += LP(fixit, sticks) - LP(pp, sticks);
        ticks += LP(fixit, uticks) - LP(pp, uticks);
        if (ticks > d)
                ticks = d;
        LP(fixit, pctcpu) = (ticks * (uint64_t)fscale) / d;
}

caddr_t 
get_process_info(struct system_info *si, struct process_select *sel,
    int compare_index)
{
        struct timespec tv;
        uint64_t t, d = 0;

        int i;
        int total_procs;
        int active_procs;
        struct kinfo_proc **prefp;
        struct kinfo_proc *pp;

        /* these are copied out of sel for speed */
        int show_idle;
        int show_system;
        int show_uid;
        int show_threads;
        char *match_command;

        show_threads = sel->threads;

        pbase = kvm_getprocs(kd,
            KERN_PROC_ALL | (show_threads ? KERN_PROC_FLAG_LWP : 0), 0, &nproc);
        if (nproc > onproc)
                pref = (struct kinfo_proc **)realloc(pref, sizeof(struct kinfo_proc *)
                    * (onproc = nproc));
        if (pref == NULL || pbase == NULL) {
                (void)fprintf(stderr, "top: Out of memory.\n");
                quit(23);
        }

        clock_gettime(CLOCK_MONOTONIC_PRECISE, &tv);
        t = (tv.tv_sec * 1000000ULL) + (tv.tv_nsec / 1000ULL);
        if (prev_pbase_time > 0 && t > prev_pbase_time)
                d = t - prev_pbase_time;

        /* get a pointer to the states summary array */
        si->procstates = process_states;

        /* set up flags which define what we are going to select */
        show_idle = sel->idle;
        show_system = sel->system;
        show_uid = sel->uid != -1;
        show_fullcmd = sel->fullcmd;
        match_command = sel->command;

        /* count up process states and get pointers to interesting procs */
        total_procs = 0;
        active_procs = 0;
        memset((char *)process_states, 0, sizeof(process_states));
        prefp = pref;
        for (pp = pbase, i = 0; i < nproc; pp++, i++) {
                /*
                 * Place pointers to each valid proc structure in pref[].
                 * Process slots that are actually in use have a non-zero
                 * status field.  Processes with P_SYSTEM set are system
                 * processes---these get ignored unless show_sysprocs is set.
                 */
                if ((show_system && (LP(pp, pid) == -1)) ||
                    (show_system || ((PP(pp, flags) & P_SYSTEM) == 0))) {
                        int lpstate = LP(pp, stat);
                        int pstate = PP(pp, stat);

                        total_procs++;
                        if (lpstate == LSRUN)
                                process_states[0]++;
                        if (pstate >= 0 && pstate < MAXPSTATES - 1)
                                process_states[pstate]++;

                        if (match_command != NULL &&
                            strstr(PP(pp, comm), match_command) == NULL) {
                                /* Command does not match */
                                continue;
                        }

                        if (show_uid && PP(pp, ruid) != (uid_t)sel->uid) {
                                /* UID does not match */
                                continue;
                        }

                        if (!show_system && LP(pp, pid) == -1) {
                                /* Don't show system processes */
                                continue;
                        }

                        /* Fix up pctcpu before show_idle test */
                        fixup_pctcpu(pp, d);

                        if (!show_idle && LP(pp, pctcpu) == 0 &&
                            lpstate != LSRUN) {
                                /* Don't show idle processes */
                                continue;
                        }

                        *prefp++ = pp;
                        active_procs++;
                }
        }

        /*
         * Save kinfo_procs for later pctcpu fixup.
         */
        if (prev_pbase_alloc < nproc) {
                prev_pbase_alloc = nproc;
                prev_pbase = realloc(prev_pbase,
                    prev_pbase_alloc * sizeof(struct kinfo_proc));
                if (prev_pbase == NULL) {
                        fprintf(stderr, "top: Out of memory.\n");
                        quit(23);
                }
        }
        prev_nproc = nproc;
        prev_pbase_time = t;
        memcpy(prev_pbase, pbase, nproc * sizeof(struct kinfo_proc));

        qsort((char *)pref, active_procs, sizeof(struct kinfo_proc *),
            (int (*)(const void *, const void *))proc_compares[compare_index]);

        /* remember active and total counts */
        si->p_total = total_procs;
        si->p_active = pref_len = active_procs;

        /* pass back a handle */
        handle.next_proc = pref;
        handle.remaining = active_procs;
        handle.show_threads = show_threads;
        return ((caddr_t) & handle);
}

char fmt[MAX_COLS];             /* static area where result is built */

char *
format_next_process(caddr_t xhandle, char *(*get_userid) (int))
{
        struct kinfo_proc *pp;
        long cputime;
        long ccputime;
        double pct;
        struct handle *hp;
        char status[16];
        int state;
        int xnice;
        char *comm;
        char cputime_fmt[10], ccputime_fmt[10];

        /* find and remember the next proc structure */
        hp = (struct handle *)xhandle;
        pp = *(hp->next_proc++);
        hp->remaining--;

        /* get the process's command name */
        if (show_fullcmd) {
                char **comm_full = kvm_getargv(kd, pp, 0);
                if (comm_full != 0)
                        comm = *comm_full;
                else
                        comm = PP(pp, comm);
        }
        else {
                comm = PP(pp, comm);
        }

        /* the actual field to display */
        char cmdfield[MAX_COLS];

        if (PP(pp, flags) & P_SYSTEM) {
                /* system process */
                snprintf(cmdfield, sizeof cmdfield, "[%s]", comm);
        } else if (hp->show_threads && PP(pp, nthreads) > 1) {
                /* display it as a thread */
                if (strcmp(PP(pp, comm), LP(pp, comm)) == 0) {
                        snprintf(cmdfield, sizeof cmdfield, "%s{%d}", comm,
                            LP(pp, tid));
                } else {
                        /* show thread name in addition to tid */
                        snprintf(cmdfield, sizeof cmdfield, "%s{%d/%s}", comm,
                            LP(pp, tid), LP(pp, comm));
                }
        } else {
                snprintf(cmdfield, sizeof cmdfield, "%s", comm);
        }
        
        /*
         * Convert the process's runtime from microseconds to seconds.  This
         * time includes the interrupt time to be in compliance with ps output.
         */
        cputime = (LP(pp, uticks) + LP(pp, sticks) + LP(pp, iticks)) / 1000000;
        ccputime = cputime + PP(pp, cru).ru_stime.tv_sec + PP(pp, cru).ru_utime.tv_sec;
        format_time(cputime, cputime_fmt, sizeof(cputime_fmt));
        format_time(ccputime, ccputime_fmt, sizeof(ccputime_fmt));

        /* calculate the base for cpu percentages */
        pct = pctdouble(LP(pp, pctcpu));

        /* generate "STATE" field */
        switch (state = LP(pp, stat)) {
        case LSRUN:
                if (LP(pp, tdflags) & TDF_RUNNING)
                        sprintf(status, "CPU%d", LP(pp, cpuid));
                else
                        strcpy(status, "RUN");
                break;
        case LSSLEEP:
                if (LP(pp, wmesg) != NULL) {
                        sprintf(status, "%.8s", LP(pp, wmesg)); /* WMESGLEN */
                        break;
                }
                /* fall through */
        default:

                if (state >= 0 &&
                    (unsigned)state < sizeof(state_abbrev) / sizeof(*state_abbrev))
                        sprintf(status, "%.6s", state_abbrev[(unsigned char)state]);
                else
                        sprintf(status, "?%5d", state);
                break;
        }

        if (PP(pp, stat) == SZOMB)
                strcpy(status, "ZOMB");

        /*
         * idle time 0 - 31 -> nice value +21 - +52 normal time      -> nice
         * value -20 - +20 real time 0 - 31 -> nice value -52 - -21 thread
         * 0 - 31 -> nice value -53 -
         */
        switch (LP(pp, rtprio.type)) {
        case RTP_PRIO_REALTIME:
                xnice = PRIO_MIN - 1 - RTP_PRIO_MAX + LP(pp, rtprio.prio);
                break;
        case RTP_PRIO_IDLE:
                xnice = PRIO_MAX + 1 + LP(pp, rtprio.prio);
                break;
        case RTP_PRIO_THREAD:
                xnice = PRIO_MIN - 1 - RTP_PRIO_MAX - LP(pp, rtprio.prio);
                break;
        default:
                xnice = PP(pp, nice);
                break;
        }

        /* format this entry */
        snprintf(fmt, sizeof(fmt),
            smp_Proc_format,
            (int)PP(pp, pid),
            namelength, namelength,
            get_userid(PP(pp, ruid)),
            (int)xnice,
            format_k(PROCSIZE(pp)),
            format_k(pagetok(VP(pp, rssize))),
            status,
            LP(pp, cpuid),
            cputime_fmt,
            ccputime_fmt,
            100.0 * pct,
            cmdlength,
            cmdfield);

        /* return the result */
        return (fmt);
}

/* comparison routines for qsort */

/*
 *  proc_compare - comparison function for "qsort"
 *      Compares the resource consumption of two processes using five
 *      distinct keys.  The keys (in descending order of importance) are:
 *      percent cpu, cpu ticks, state, resident set size, total virtual
 *      memory usage.  The process states are ordered as follows (from least
 *      to most important):  WAIT, zombie, sleep, stop, start, run.  The
 *      array declaration below maps a process state index into a number
 *      that reflects this ordering.
 */

static unsigned char sorted_state[] =
{
        0,                      /* not used              */
        3,                      /* sleep                 */
        1,                      /* ABANDONED (WAIT)      */
        6,                      /* run                   */
        5,                      /* start                 */
        2,                      /* zombie                */
        4                       /* stop                  */
};


#define ORDERKEY_PCTCPU \
  if (lresult = (long) LP(p2, pctcpu) - (long) LP(p1, pctcpu), \
     (result = lresult > 0 ? 1 : lresult < 0 ? -1 : 0) == 0)

#define CPTICKS(p)      (LP(p, uticks) + LP(p, sticks) + LP(p, iticks))

#define ORDERKEY_CPTICKS \
  if ((result = CPTICKS(p2) > CPTICKS(p1) ? 1 : \
                CPTICKS(p2) < CPTICKS(p1) ? -1 : 0) == 0)

#define CTIME(p)        (((LP(p, uticks) + LP(p, sticks) + LP(p, iticks))/1000000) + \
  PP(p, cru).ru_stime.tv_sec + PP(p, cru).ru_utime.tv_sec)

#define ORDERKEY_CTIME \
   if ((result = CTIME(p2) > CTIME(p1) ? 1 : \
                CTIME(p2) < CTIME(p1) ? -1 : 0) == 0)

#define ORDERKEY_STATE \
  if ((result = sorted_state[(unsigned char) PP(p2, stat)] - \
                sorted_state[(unsigned char) PP(p1, stat)]) == 0)

#define ORDERKEY_PRIO \
  if ((result = LP(p2, prio) - LP(p1, prio)) == 0)

#define ORDERKEY_KTHREADS \
  if ((result = (LP(p1, pid) == 0) - (LP(p2, pid) == 0)) == 0)

#define ORDERKEY_KTHREADS_PRIO \
  if ((result = LP(p2, tdprio) - LP(p1, tdprio)) == 0)

#define ORDERKEY_RSSIZE \
  if ((result = VP(p2, rssize) - VP(p1, rssize)) == 0)

#define ORDERKEY_MEM \
  if ( (result = PROCSIZE(p2) - PROCSIZE(p1)) == 0 )

#define ORDERKEY_PID \
  if ( (result = PP(p1, pid) - PP(p2, pid)) == 0)

#define ORDERKEY_PRSSIZE \
  if((result = VP(p2, prssize) - VP(p1, prssize)) == 0)

static __inline int
orderkey_kernidle(const struct kinfo_proc *p1, const struct kinfo_proc *p2)
{
        int p1_kidle = 0, p2_kidle = 0;

        if (LP(p1, pid) == -1 && PP(p1, stat) == SIDL)
                p1_kidle = 1;
        if (LP(p2, pid) == -1 && PP(p2, stat) == SIDL)
                p2_kidle = 1;

        if (!p2_kidle && p1_kidle)
                return 1;
        if (p2_kidle && !p1_kidle)
                return -1;
        return 0;
}

#define ORDERKEY_KIDLE  if ((result = orderkey_kernidle(p1, p2)) == 0)

/* compare_cpu - the comparison function for sorting by cpu percentage */

int
proc_compare(struct kinfo_proc **pp1, struct kinfo_proc **pp2)
{
        struct kinfo_proc *p1;
        struct kinfo_proc *p2;
        int result;
        pctcpu lresult;

        /* remove one level of indirection */
        p1 = *(struct kinfo_proc **) pp1;
        p2 = *(struct kinfo_proc **) pp2;

        ORDERKEY_KIDLE
        ORDERKEY_PCTCPU
        ORDERKEY_CPTICKS
        ORDERKEY_STATE
        ORDERKEY_PRIO
        ORDERKEY_RSSIZE
        ORDERKEY_MEM
        {} 
        
        return (result);
}

/* compare_size - the comparison function for sorting by total memory usage */

int
compare_size(struct kinfo_proc **pp1, struct kinfo_proc **pp2)
{
        struct kinfo_proc *p1;
        struct kinfo_proc *p2;
        int result;
        pctcpu lresult;

        /* remove one level of indirection */
        p1 = *(struct kinfo_proc **) pp1;
        p2 = *(struct kinfo_proc **) pp2;

        ORDERKEY_MEM
        ORDERKEY_RSSIZE
        ORDERKEY_KIDLE
        ORDERKEY_PCTCPU
        ORDERKEY_CPTICKS
        ORDERKEY_STATE
        ORDERKEY_PRIO
        {}

        return (result);
}

/* compare_res - the comparison function for sorting by resident set size */

int
compare_res(struct kinfo_proc **pp1, struct kinfo_proc **pp2)
{
        struct kinfo_proc *p1;
        struct kinfo_proc *p2;
        int result;
        pctcpu lresult;

        /* remove one level of indirection */
        p1 = *(struct kinfo_proc **) pp1;
        p2 = *(struct kinfo_proc **) pp2;

        ORDERKEY_RSSIZE
        ORDERKEY_MEM
        ORDERKEY_KIDLE
        ORDERKEY_PCTCPU
        ORDERKEY_CPTICKS
        ORDERKEY_STATE
        ORDERKEY_PRIO
        {}

        return (result);
}

/* compare_pres - the comparison function for sorting by proportional resident set size */

int
compare_pres(struct kinfo_proc **pp1, struct kinfo_proc **pp2)
{
        struct kinfo_proc *p1;
        struct kinfo_proc *p2;
        int result;
        pctcpu lresult;

        /* remove one level of indirection */
        p1 = *(struct kinfo_proc **) pp1;
        p2 = *(struct kinfo_proc **) pp2;

        ORDERKEY_PRSSIZE
        ORDERKEY_RSSIZE
        ORDERKEY_MEM
        ORDERKEY_KIDLE
        ORDERKEY_PCTCPU
        ORDERKEY_CPTICKS
        ORDERKEY_STATE
        ORDERKEY_PRIO
        {}

        return (result);
}

/* compare_time - the comparison function for sorting by total cpu time */

int
compare_time(struct kinfo_proc **pp1, struct kinfo_proc **pp2)
{
        struct kinfo_proc *p1;
        struct kinfo_proc *p2;
        int result;
        pctcpu lresult;

        /* remove one level of indirection */
        p1 = *(struct kinfo_proc **) pp1;
        p2 = *(struct kinfo_proc **) pp2;

        ORDERKEY_KIDLE
        ORDERKEY_CPTICKS
        ORDERKEY_PCTCPU
        ORDERKEY_KTHREADS
        ORDERKEY_KTHREADS_PRIO
        ORDERKEY_STATE
        ORDERKEY_PRIO
        ORDERKEY_RSSIZE
        ORDERKEY_MEM
        {}

        return (result);
}

int
compare_ctime(struct kinfo_proc **pp1, struct kinfo_proc **pp2)
{
        struct kinfo_proc *p1;
        struct kinfo_proc *p2;
        int result;
        pctcpu lresult;
        
        /* remove one level of indirection */
        p1 = *(struct kinfo_proc **) pp1;
        p2 = *(struct kinfo_proc **) pp2;

        ORDERKEY_KIDLE
        ORDERKEY_CTIME
        ORDERKEY_PCTCPU
        ORDERKEY_KTHREADS
        ORDERKEY_KTHREADS_PRIO
        ORDERKEY_STATE
        ORDERKEY_PRIO
        ORDERKEY_RSSIZE
        ORDERKEY_MEM
        {}
        
        return (result);
}

/* compare_prio - the comparison function for sorting by cpu percentage */

int
compare_prio(struct kinfo_proc **pp1, struct kinfo_proc **pp2)
{
        struct kinfo_proc *p1;
        struct kinfo_proc *p2;
        int result;
        pctcpu lresult;

        /* remove one level of indirection */
        p1 = *(struct kinfo_proc **) pp1;
        p2 = *(struct kinfo_proc **) pp2;

        ORDERKEY_KTHREADS
        ORDERKEY_KTHREADS_PRIO
        ORDERKEY_PRIO
        ORDERKEY_KIDLE
        ORDERKEY_CPTICKS
        ORDERKEY_PCTCPU
        ORDERKEY_STATE
        ORDERKEY_RSSIZE
        ORDERKEY_MEM
        {}

        return (result);
}

int
compare_thr(struct kinfo_proc **pp1, struct kinfo_proc **pp2)
{
        struct kinfo_proc *p1;
        struct kinfo_proc *p2;
        int result;
        pctcpu lresult;

        /* remove one level of indirection */
        p1 = *(struct kinfo_proc **)pp1;
        p2 = *(struct kinfo_proc **)pp2;

        ORDERKEY_KTHREADS
        ORDERKEY_KTHREADS_PRIO
        ORDERKEY_KIDLE
        ORDERKEY_CPTICKS
        ORDERKEY_PCTCPU
        ORDERKEY_STATE
        ORDERKEY_RSSIZE
        ORDERKEY_MEM
        {}

        return (result);
}

/* compare_pid - the comparison function for sorting by process id */

int
compare_pid(struct kinfo_proc **pp1, struct kinfo_proc **pp2)
{
        struct kinfo_proc *p1;
        struct kinfo_proc *p2;
        int result;

        /* remove one level of indirection */
        p1 = *(struct kinfo_proc **) pp1;
        p2 = *(struct kinfo_proc **) pp2;
        
        ORDERKEY_PID
        ;
        
        return(result);
}

/*
 * proc_owner(pid) - returns the uid that owns process "pid", or -1 if
 *              the process does not exist.
 *              It is EXTREMLY IMPORTANT that this function work correctly.
 *              If top runs setuid root (as in SVR4), then this function
 *              is the only thing that stands in the way of a serious
 *              security problem.  It validates requests for the "kill"
 *              and "renice" commands.
 */

int
proc_owner(int pid)
{
        int xcnt;
        struct kinfo_proc **prefp;
        struct kinfo_proc *pp;

        prefp = pref;
        xcnt = pref_len;
        while (--xcnt >= 0) {
                pp = *prefp++;
                if (PP(pp, pid) == (pid_t) pid) {
                        return ((int)PP(pp, ruid));
                }
        }
        return (-1);
}


/*
 * swapmode is based on a program called swapinfo written
 * by Kevin Lahey <kml@rokkaku.atl.ga.us>.
 */
int
swapmode(int *retavail, int *retfree)
{
        int n;
        int pagesize = getpagesize();
        struct kvm_swap swapary[1];

        *retavail = 0;
        *retfree = 0;

#define CONVERT(v)      ((quad_t)(v) * pagesize / 1024)

        n = kvm_getswapinfo(kd, swapary, 1, 0);
        if (n < 0 || swapary[0].ksw_total == 0)
                return (0);

        *retavail = CONVERT(swapary[0].ksw_total);
        *retfree = CONVERT(swapary[0].ksw_total - swapary[0].ksw_used);

        n = (int)((double)swapary[0].ksw_used * 100.0 /
            (double)swapary[0].ksw_total);
        return (n);
}