MFC r315280 r315287

When the RTC is adjusted, reevaluate absolute sleep times based on the RTC

POSIX 2008 says this about clock_settime(2):

    If the value of the CLOCK_REALTIME clock is set via clock_settime(),
    the new value of the clock shall be used to determine the time
    of expiration for absolute time services based upon the
    CLOCK_REALTIME clock.  This applies to the time at which armed
    absolute timers expire.  If the absolute time requested at the
    invocation of such a time service is before the new value of
    the clock, the time service shall expire immediately as if the
    clock had reached the requested time normally.

    Setting the value of the CLOCK_REALTIME clock via clock_settime()
    shall have no effect on threads that are blocked waiting for
    a relative time service based upon this clock, including the
    nanosleep() function; nor on the expiration of relative timers
    based upon this clock.  Consequently, these time services shall
    expire when the requested relative interval elapses, independently
    of the new or old value of the clock.

When the real-time clock is adjusted, such as by clock_settime(3),
wake any threads sleeping until an absolute real-clock time.
Such a sleep is indicated by a non-zero td_rtcgen.  The sleep functions
will set that field to zero and return zero to tell the caller
to reevaluate its sleep duration based on the new value of the clock.

At present, this affects the following functions:

    pthread_cond_timedwait(3)
    pthread_mutex_timedlock(3)
    pthread_rwlock_timedrdlock(3)
    pthread_rwlock_timedwrlock(3)
    sem_timedwait(3)
    sem_clockwait_np(3)

I'm working on adding clock_nanosleep(2), which will also be affected.

Reported by:	Sebastian Huber <sebastian.huber@embedded-brains.de>
Relnotes:	yes
Sponsored by:	Dell EMC

1 files changed, 39 insertions, 9 deletions

diff --git a/sys/kern/kern_umtx.c b/sys/kern/kern_umtx.c
index 5accaf7..c33d00c 100644
--- a/sys/kern/kern_umtx.c
+++ b/sys/kern/kern_umtx.c
@@ -59,6 +59,7 @@ __FBSDID("$FreeBSD$");
 #include <sys/sysproto.h>
 #include <sys/syscallsubr.h>
 #include <sys/taskqueue.h>
+#include <sys/time.h>
 #include <sys/eventhandler.h>
 #include <sys/umtx.h>
 
@@ -70,6 +71,7 @@ __FBSDID("$FreeBSD$");
 #include <vm/vm_map.h>
 #include <vm/vm_object.h>
 
+#include <machine/atomic.h>
 #include <machine/cpu.h>
 
 #ifdef COMPAT_FREEBSD32
@@ -208,6 +210,7 @@ struct umtxq_chain {
 
 struct abs_timeout {
 	int clockid;
+	bool is_abs_real;	/* TIMER_ABSTIME && CLOCK_REALTIME* */
 	struct timespec cur;
 	struct timespec end;
 };
@@ -255,6 +258,8 @@ SYSCTL_LONG(_debug_umtx, OID_AUTO, max_length, CTLFLAG_RD, &max_length, 0, "max_
 static SYSCTL_NODE(_debug_umtx, OID_AUTO, chains, CTLFLAG_RD, 0, "umtx chain stats");
 #endif
 
+static void abs_timeout_update(struct abs_timeout *timo);
+
 static void umtx_shm_init(void);
 static void umtxq_sysinit(void *);
 static void umtxq_hash(struct umtx_key *key);
@@ -770,12 +775,22 @@ abs_timeout_init(struct abs_timeout *timo, int clockid, int absolute,
 
 	timo->clockid = clockid;
 	if (!absolute) {
-		kern_clock_gettime(curthread, clockid, &timo->end);
-		timo->cur = timo->end;
+		timo->is_abs_real = false;
+		abs_timeout_update(timo);
+		timo->end = timo->cur;
 		timespecadd(&timo->end, timeout);
 	} else {
 		timo->end = *timeout;
-		kern_clock_gettime(curthread, clockid, &timo->cur);
+		timo->is_abs_real = clockid == CLOCK_REALTIME ||
+		    clockid == CLOCK_REALTIME_FAST ||
+		    clockid == CLOCK_REALTIME_PRECISE;
+		/*
+		 * If is_abs_real, umtxq_sleep will read the clock
+		 * after setting td_rtcgen; otherwise, read it here.
+		 */
+		if (!timo->is_abs_real) {
+			abs_timeout_update(timo);
+		}
 	}
 }
 
@@ -829,26 +844,41 @@ umtxq_sleep(struct umtx_q *uq, const char *wmesg, struct abs_timeout *abstime)
 	struct umtxq_chain *uc;
 	int error, timo;
 
+	if (abstime != NULL && abstime->is_abs_real) {
+		curthread->td_rtcgen = atomic_load_acq_int(&rtc_generation);
+		abs_timeout_update(abstime);
+	}
+
 	uc = umtxq_getchain(&uq->uq_key);
 	UMTXQ_LOCKED_ASSERT(uc);
 	for (;;) {
-		if (!(uq->uq_flags & UQF_UMTXQ))
-			return (0);
+		if (!(uq->uq_flags & UQF_UMTXQ)) {
+			error = 0;
+			break;
+		}
 		if (abstime != NULL) {
 			timo = abs_timeout_gethz(abstime);
-			if (timo < 0)
-				return (ETIMEDOUT);
+			if (timo < 0) {
+				error = ETIMEDOUT;
+				break;
+			}
 		} else
 			timo = 0;
 		error = msleep(uq, &uc->uc_lock, PCATCH | PDROP, wmesg, timo);
-		if (error != EWOULDBLOCK) {
+		if (error == EINTR || error == ERESTART) {
 			umtxq_lock(&uq->uq_key);
 			break;
 		}
-		if (abstime != NULL)
+		if (abstime != NULL) {
+			if (abstime->is_abs_real)
+				curthread->td_rtcgen =
+				    atomic_load_acq_int(&rtc_generation);
 			abs_timeout_update(abstime);
+		}
 		umtxq_lock(&uq->uq_key);
 	}
+
+	curthread->td_rtcgen = 0;
 	return (error);
 }