#include <sys/kauth.h>
#include <sys/vnode.h>
#include <sys/time.h>
+#include <sys/priv.h>
#include <sys/mount_internal.h>
#include <sys/sysproto.h>
#include <sys/signalvar.h>
+#include <sys/protosw.h> /* for net_uptime2timeval() */
#include <kern/clock.h>
#include <kern/task.h>
#if CONFIG_MACF
#include <security/mac_framework.h>
#endif
+#include <IOKit/IOBSD.h>
+#include <sys/time.h>
#define HZ 100 /* XXX */
static void setthetime(
struct timeval *tv);
-void time_zone_slock_init(void) __attribute__((section("__TEXT, initcode")));
+void time_zone_slock_init(void);
+static boolean_t timeval_fixusec(struct timeval *t1);
-/*
+/*
* Time of day and interval timer support.
*
* These routines provide the kernel entry points to get and set
/* ARGSUSED */
int
gettimeofday(
-__unused struct proc *p,
- struct gettimeofday_args *uap,
- int32_t *retval)
+ struct proc *p,
+ struct gettimeofday_args *uap,
+ __unused int32_t *retval)
{
int error = 0;
struct timezone ltz; /* local copy */
+ clock_sec_t secs;
+ clock_usec_t usecs;
+ uint64_t mach_time;
- if (uap->tp) {
- clock_sec_t secs;
- clock_usec_t usecs;
+ if (uap->tp || uap->mach_absolute_time) {
+ clock_gettimeofday_and_absolute_time(&secs, &usecs, &mach_time);
+ }
- clock_gettimeofday(&secs, &usecs);
- retval[0] = secs;
- retval[1] = usecs;
+ if (uap->tp) {
+ /* Casting secs through a uint32_t to match arm64 commpage */
+ if (IS_64BIT_PROCESS(p)) {
+ struct user64_timeval user_atv = {};
+ user_atv.tv_sec = (uint32_t)secs;
+ user_atv.tv_usec = usecs;
+ error = copyout(&user_atv, uap->tp, sizeof(user_atv));
+ } else {
+ struct user32_timeval user_atv = {};
+ user_atv.tv_sec = (uint32_t)secs;
+ user_atv.tv_usec = usecs;
+ error = copyout(&user_atv, uap->tp, sizeof(user_atv));
+ }
+ if (error) {
+ return error;
+ }
}
-
+
if (uap->tzp) {
lck_spin_lock(tz_slock);
ltz = tz;
lck_spin_unlock(tz_slock);
- error = copyout((caddr_t)<z, CAST_USER_ADDR_T(uap->tzp), sizeof (tz));
+ error = copyout((caddr_t)<z, CAST_USER_ADDR_T(uap->tzp), sizeof(tz));
}
- return (error);
+ if (error == 0 && uap->mach_absolute_time) {
+ error = copyout(&mach_time, uap->mach_absolute_time, sizeof(mach_time));
+ }
+
+ return error;
}
/*
bzero(&atv, sizeof(atv));
+ /* Check that this task is entitled to set the time or it is root */
+ if (!IOTaskHasEntitlement(current_task(), SETTIME_ENTITLEMENT)) {
+
#if CONFIG_MACF
- error = mac_system_check_settime(kauth_cred_get());
- if (error)
- return (error);
+ error = mac_system_check_settime(kauth_cred_get());
+ if (error)
+ return (error);
#endif
#ifndef CONFIG_EMBEDDED
- if ((error = suser(kauth_cred_get(), &p->p_acflag)))
- return (error);
+ if ((error = suser(kauth_cred_get(), &p->p_acflag)))
+ return (error);
#endif
+ }
+
/* Verify all parameters before changing time */
if (uap->tv) {
if (IS_64BIT_PROCESS(p)) {
if (uap->tzp && (error = copyin(uap->tzp, (caddr_t)&atz, sizeof(atz))))
return (error);
if (uap->tv) {
- timevalfix(&atv);
- if (atv.tv_sec < 0 || (atv.tv_sec == 0 && atv.tv_usec < 0))
+ /* only positive values of sec/usec are accepted */
+ if (atv.tv_sec < 0 || atv.tv_usec < 0)
+ return (EPERM);
+ if (!timeval_fixusec(&atv))
return (EPERM);
setthetime(&atv);
}
clock_set_calendar_microtime(tv->tv_sec, tv->tv_usec);
}
-/*
- * XXX Y2038 bug because of clock_adjtime() first argument
- */
-/* ARGSUSED */
-int
-adjtime(struct proc *p, struct adjtime_args *uap, __unused int32_t *retval)
-{
- struct timeval atv;
- int error;
-
-#if CONFIG_MACF
- error = mac_system_check_settime(kauth_cred_get());
- if (error)
- return (error);
-#endif
- if ((error = suser(kauth_cred_get(), &p->p_acflag)))
- return (error);
- if (IS_64BIT_PROCESS(p)) {
- struct user64_timeval user_atv;
- error = copyin(uap->delta, &user_atv, sizeof(user_atv));
- atv.tv_sec = user_atv.tv_sec;
- atv.tv_usec = user_atv.tv_usec;
- } else {
- struct user32_timeval user_atv;
- error = copyin(uap->delta, &user_atv, sizeof(user_atv));
- atv.tv_sec = user_atv.tv_sec;
- atv.tv_usec = user_atv.tv_usec;
- }
- if (error)
- return (error);
-
- /*
- * Compute the total correction and the rate at which to apply it.
- */
- clock_adjtime(&atv.tv_sec, &atv.tv_usec);
-
- if (uap->olddelta) {
- if (IS_64BIT_PROCESS(p)) {
- struct user64_timeval user_atv;
- user_atv.tv_sec = atv.tv_sec;
- user_atv.tv_usec = atv.tv_usec;
- error = copyout(&user_atv, uap->olddelta, sizeof(user_atv));
- } else {
- struct user32_timeval user_atv;
- user_atv.tv_sec = atv.tv_sec;
- user_atv.tv_usec = atv.tv_usec;
- error = copyout(&user_atv, uap->olddelta, sizeof(user_atv));
- }
- }
-
- return (0);
-}
-
/*
* Verify the calendar value. If negative,
* reset to zero (the epoch).
return (secs);
}
+void
+boottime_timeval(struct timeval *tv)
+{
+ clock_sec_t secs;
+ clock_usec_t microsecs;
+
+ clock_get_boottime_microtime(&secs, µsecs);
+
+ tv->tv_sec = secs;
+ tv->tv_usec = microsecs;
+}
+
/*
* Get value of an interval timer. The process virtual and
* profiling virtual time timers are kept internally in the
if (IS_64BIT_PROCESS(p)) {
struct user64_itimerval user_itv;
+ bzero(&user_itv, sizeof (user_itv));
user_itv.it_interval.tv_sec = aitv.it_interval.tv_sec;
user_itv.it_interval.tv_usec = aitv.it_interval.tv_usec;
user_itv.it_value.tv_sec = aitv.it_value.tv_sec;
return (copyout((caddr_t)&user_itv, uap->itv, sizeof (user_itv)));
} else {
struct user32_itimerval user_itv;
+ bzero(&user_itv, sizeof (user_itv));
user_itv.it_interval.tv_sec = aitv.it_interval.tv_sec;
user_itv.it_interval.tv_usec = aitv.it_interval.tv_usec;
user_itv.it_value.tv_sec = aitv.it_value.tv_sec;
microuptime(&p->p_rtime);
timevaladd(&p->p_rtime, &aitv.it_value);
p->p_realtimer = aitv;
- if (!thread_call_enter_delayed(p->p_rcall, tvtoabstime(&p->p_rtime)))
+ if (!thread_call_enter_delayed_with_leeway(p->p_rcall, NULL,
+ tvtoabstime(&p->p_rtime), 0, THREAD_CALL_DELAY_USER_NORMAL))
p->p_ractive++;
} else {
timerclear(&p->p_rtime);
*/
void
realitexpire(
- struct proc *p)
+ struct proc *p)
{
struct proc *r;
- struct timeval t;
+ struct timeval t;
r = proc_find(p->p_pid);
proc_spinlock(p);
+ assert(p->p_ractive > 0);
+
if (--p->p_ractive > 0 || r != p) {
+ /*
+ * bail, because either proc is exiting
+ * or there's another active thread call
+ */
proc_spinunlock(p);
if (r != NULL)
proc_rele(r);
return;
}
-
+
if (!timerisset(&p->p_realtimer.it_interval)) {
+ /*
+ * p_realtimer was cleared while this call was pending,
+ * send one last SIGALRM, but don't re-arm
+ */
timerclear(&p->p_rtime);
proc_spinunlock(p);
return;
}
+ proc_spinunlock(p);
+
+ /*
+ * Send the signal before re-arming the next thread call,
+ * so in case psignal blocks, we won't create yet another thread call.
+ */
+
+ psignal(p, SIGALRM);
+
+ proc_spinlock(p);
+
+ /* Should we still re-arm the next thread call? */
+ if (!timerisset(&p->p_realtimer.it_interval)) {
+ timerclear(&p->p_rtime);
+ proc_spinunlock(p);
+
+ proc_rele(p);
+ return;
+ }
+
microuptime(&t);
timevaladd(&p->p_rtime, &p->p_realtimer.it_interval);
+
if (timercmp(&p->p_rtime, &t, <=)) {
if ((p->p_rtime.tv_sec + 2) >= t.tv_sec) {
for (;;) {
if (timercmp(&p->p_rtime, &t, >))
break;
}
- }
- else {
+ } else {
p->p_rtime = p->p_realtimer.it_interval;
timevaladd(&p->p_rtime, &t);
}
}
- if (!thread_call_enter_delayed(p->p_rcall, tvtoabstime(&p->p_rtime)))
+ assert(p->p_rcall != NULL);
+
+ if (!thread_call_enter_delayed_with_leeway(p->p_rcall, NULL, tvtoabstime(&p->p_rtime), 0,
+ THREAD_CALL_DELAY_USER_NORMAL)) {
p->p_ractive++;
+ }
+
proc_spinunlock(p);
- psignal(p, SIGALRM);
proc_rele(p);
}
+/*
+ * Called once in proc_exit to clean up after an armed or pending realitexpire
+ *
+ * This will only be called after the proc refcount is drained,
+ * so realitexpire cannot be currently holding a proc ref.
+ * i.e. it will/has gotten PROC_NULL from proc_find.
+ */
+void
+proc_free_realitimer(proc_t p)
+{
+ proc_spinlock(p);
+
+ assert(p->p_rcall != NULL);
+ assert(p->p_refcount == 0);
+
+ timerclear(&p->p_realtimer.it_interval);
+
+ if (thread_call_cancel(p->p_rcall)) {
+ assert(p->p_ractive > 0);
+ p->p_ractive--;
+ }
+
+ while (p->p_ractive > 0) {
+ proc_spinunlock(p);
+
+ delay(1);
+
+ proc_spinlock(p);
+ }
+
+ thread_call_t call = p->p_rcall;
+ p->p_rcall = NULL;
+
+ proc_spinunlock(p);
+
+ thread_call_free(call);
+}
+
/*
* Check that a proposed value to load into the .it_value or
* .it_interval part of an interval timer is acceptable.
return (0);
}
+int
+timespec_is_valid(const struct timespec *ts)
+{
+ /* The INT32_MAX limit ensures the timespec is safe for clock_*() functions
+ * which accept 32-bit ints. */
+ if (ts->tv_sec < 0 || ts->tv_sec > INT32_MAX ||
+ ts->tv_nsec < 0 || (unsigned long long)ts->tv_nsec > NSEC_PER_SEC) {
+ return 0;
+ }
+ return 1;
+}
+
/*
* Decrement an interval timer by a specified number
* of microseconds, which must be less than a second,
}
}
+static boolean_t
+timeval_fixusec(
+ struct timeval *t1)
+{
+ assert(t1->tv_usec >= 0);
+ assert(t1->tv_sec >= 0);
+
+ if (t1->tv_usec >= 1000000) {
+ if (os_add_overflow(t1->tv_sec, t1->tv_usec / 1000000, &t1->tv_sec))
+ return FALSE;
+ t1->tv_usec = t1->tv_usec % 1000000;
+ }
+
+ return TRUE;
+}
+
/*
* Return the best possible estimate of the time in the timeval
* to which tvp points.
tvp->tv_usec = tv_usec;
}
+void
+microtime_with_abstime(
+ struct timeval *tvp, uint64_t *abstime)
+{
+ clock_sec_t tv_sec;
+ clock_usec_t tv_usec;
+
+ clock_get_calendar_absolute_and_microtime(&tv_sec, &tv_usec, abstime);
+
+ tvp->tv_sec = tv_sec;
+ tvp->tv_usec = tv_usec;
+}
+
void
microuptime(
struct timeval *tvp)
return (result + usresult);
}
+
+uint64_t
+tstoabstime(struct timespec *ts)
+{
+ uint64_t abstime_s, abstime_ns;
+ clock_interval_to_absolutetime_interval(ts->tv_sec, NSEC_PER_SEC, &abstime_s);
+ clock_interval_to_absolutetime_interval(ts->tv_nsec, 1, &abstime_ns);
+ return abstime_s + abstime_ns;
+}
+
+#if NETWORKING
+/*
+ * ratecheck(): simple time-based rate-limit checking.
+ */
+int
+ratecheck(struct timeval *lasttime, const struct timeval *mininterval)
+{
+ struct timeval tv, delta;
+ int rv = 0;
+
+ net_uptime2timeval(&tv);
+ delta = tv;
+ timevalsub(&delta, lasttime);
+
+ /*
+ * check for 0,0 is so that the message will be seen at least once,
+ * even if interval is huge.
+ */
+ if (timevalcmp(&delta, mininterval, >=) ||
+ (lasttime->tv_sec == 0 && lasttime->tv_usec == 0)) {
+ *lasttime = tv;
+ rv = 1;
+ }
+
+ return (rv);
+}
+
+/*
+ * ppsratecheck(): packets (or events) per second limitation.
+ */
+int
+ppsratecheck(struct timeval *lasttime, int *curpps, int maxpps)
+{
+ struct timeval tv, delta;
+ int rv;
+
+ net_uptime2timeval(&tv);
+
+ timersub(&tv, lasttime, &delta);
+
+ /*
+ * Check for 0,0 so that the message will be seen at least once.
+ * If more than one second has passed since the last update of
+ * lasttime, reset the counter.
+ *
+ * we do increment *curpps even in *curpps < maxpps case, as some may
+ * try to use *curpps for stat purposes as well.
+ */
+ if ((lasttime->tv_sec == 0 && lasttime->tv_usec == 0) ||
+ delta.tv_sec >= 1) {
+ *lasttime = tv;
+ *curpps = 0;
+ rv = 1;
+ } else if (maxpps < 0)
+ rv = 1;
+ else if (*curpps < maxpps)
+ rv = 1;
+ else
+ rv = 0;
+
+#if 1 /* DIAGNOSTIC? */
+ /* be careful about wrap-around */
+ if (*curpps + 1 > 0)
+ *curpps = *curpps + 1;
+#else
+ /*
+ * assume that there's not too many calls to this function.
+ * not sure if the assumption holds, as it depends on *caller's*
+ * behavior, not the behavior of this function.
+ * IMHO it is wrong to make assumption on the caller's behavior,
+ * so the above #if is #if 1, not #ifdef DIAGNOSTIC.
+ */
+ *curpps = *curpps + 1;
+#endif
+
+ return (rv);
+}
+#endif /* NETWORKING */
+
void
time_zone_slock_init(void)
{
/* Allocate the spin lock */
tz_slock = lck_spin_alloc_init(tz_slock_grp, tz_slock_attr);
}
+