diff --git a/include/kernel.h b/include/kernel.h
index 94bf67313c7..def4de8a965 100644
--- a/include/kernel.h
+++ b/include/kernel.h
@@ -1606,6 +1606,32 @@ const char *k_thread_state_str(k_tid_t thread_id);
  */
 #define K_FOREVER Z_FOREVER
 
+/**
+ * @brief Generates an absolute/uptime timeout value in ticks
+ *
+ * This macro generates a timeout delay that represents an expiration
+ * at the absolute uptime value specified, in ticks.  That is, the
+ * timeout will expire immediately after the system uptime reaches the
+ * specified tick count.
+ *
+ * @param t Tick uptime value
+ * @return Timeout delay value
+ */
+#define K_TIMEOUT_ABS_TICKS(t) Z_TIMEOUT_TICKS(Z_TICK_ABS(MAX(t, 0)))
+
+/**
+ * @brief Generates an absolute/uptime timeout value in ms
+ *
+ * This macro generates a timeout delay that represents an expiration
+ * at the absolute uptime value specified, in milliseconds.  That is,
+ * the timeout will expire immediately after the system uptime reaches
+ * the specified tick count.
+ *
+ * @param t Millisecond uptime value
+ * @return Timeout delay value
+ */
+#define K_TIMEOUT_ABS_MS(t) K_TIMEOUT_ABS_TICKS(k_ms_to_ticks_ceil64(t))
+
 /**
  * @}
  */
diff --git a/include/sys_clock.h b/include/sys_clock.h
index 69b73579dee..43d5277b98b 100644
--- a/include/sys_clock.h
+++ b/include/sys_clock.h
@@ -77,6 +77,16 @@ typedef struct {
 #define Z_TIMEOUT_NS(t) Z_TIMEOUT_TICKS(k_ns_to_ticks_ceil32(MAX(t, 0)))
 #define Z_TIMEOUT_CYC(t) Z_TIMEOUT_TICKS(k_cyc_to_ticks_ceil32(MAX(t, 0)))
 
+/* Converts between absolute timeout expiration values (packed into
+ * the negative space below K_TICKS_FOREVER) and (non-negative) delta
+ * timeout values.  If the result of Z_TICK_ABS(t) is >= 0, then the
+ * value was an absolute timeout with the returend expiration time.
+ * Note that this macro is bidirectional: Z_TICK_ABS(Z_TICK_ABS(t)) ==
+ * t for all inputs, and that the representation of K_TICKS_FOREVER is
+ * the same value in both spaces!  Clever, huh?
+ */
+#define Z_TICK_ABS(t) (K_TICKS_FOREVER - 1 - (t))
+
 #else
 
 /* Legacy timeout API */
diff --git a/kernel/Kconfig b/kernel/Kconfig
index 5261b9c0552..712faf52a0b 100644
--- a/kernel/Kconfig
+++ b/kernel/Kconfig
@@ -587,7 +587,9 @@ config TIMEOUT_64BIT
 	  When this option is true, the k_ticks_t values passed to
 	  kernel APIs will be a 64 bit quantity, allowing the use of
 	  larger values (and higher precision tick rates) without fear
-	  of overflowing the 32 bit word.
+	  of overflowing the 32 bit word.  This feature also gates the
+	  availability of absolute timeout values (which require the
+	  extra precision).
 
 config XIP
 	bool "Execute in place"
diff --git a/kernel/timeout.c b/kernel/timeout.c
index 6f8efa4a76e..5d2b47921dc 100644
--- a/kernel/timeout.c
+++ b/kernel/timeout.c
@@ -90,6 +90,10 @@ void z_add_timeout(struct _timeout *to, _timeout_func_t fn,
 	k_ticks_t ticks = timeout;
 #else
 	k_ticks_t ticks = timeout.ticks + 1;
+
+	if (IS_ENABLED(CONFIG_TIMEOUT_64BIT) && Z_TICK_ABS(ticks) >= 0) {
+		ticks = Z_TICK_ABS(ticks) - (curr_tick + elapsed());
+	}
 #endif
 
 	__ASSERT(!sys_dnode_is_linked(&to->node), "");
@@ -257,7 +261,9 @@ static inline s64_t z_vrfy_k_uptime_get(void)
 #endif
 
 /* Returns the uptime expiration (relative to an unlocked "now"!) of a
- * timeout object.
+ * timeout object.  When used correctly, this should be called once,
+ * synchronously with the user passing a new timeout value.  It should
+ * not be used iteratively to adjust a timeout.
  */
 u64_t z_timeout_end_calc(k_timeout_t timeout)
 {
@@ -273,6 +279,10 @@ u64_t z_timeout_end_calc(k_timeout_t timeout)
 	dt = k_ms_to_ticks_ceil32(timeout);
 #else
 	dt = timeout.ticks;
+
+	if (IS_ENABLED(CONFIG_TIMEOUT_64BIT) && Z_TICK_ABS(dt) >= 0) {
+		return Z_TICK_ABS(dt);
+	}
 #endif
 	return z_tick_get() + MAX(1, dt);
 }
diff --git a/kernel/timer.c b/kernel/timer.c
index bcaec8a3ff1..4db44d447c9 100644
--- a/kernel/timer.c
+++ b/kernel/timer.c
@@ -127,7 +127,9 @@ void z_impl_k_timer_start(struct k_timer *timer, k_timeout_t duration,
 	 * timer_api test relies on this behavior.
 	 */
 	period.ticks = MAX(period.ticks - 1, 0);
-	duration.ticks = MAX(duration.ticks - 1, 0);
+	if (Z_TICK_ABS(duration.ticks) < 0) {
+		duration.ticks = MAX(duration.ticks - 1, 0);
+	}
 #endif
 
 	(void)z_abort_timeout(&timer->timeout);
diff --git a/tests/kernel/timer/timer_api/src/main.c b/tests/kernel/timer/timer_api/src/main.c
index 8a2dea5a617..a754dd0bbca 100644
--- a/tests/kernel/timer/timer_api/src/main.c
+++ b/tests/kernel/timer/timer_api/src/main.c
@@ -518,14 +518,21 @@ void test_timer_user_data(void)
  * k_timer_remaining_get()
  */
 
-void test_timer_remaining_get(void)
+void test_timer_remaining(void)
 {
-	u32_t remaining;
+	u32_t dur_ticks = k_ms_to_ticks_ceil32(DURATION);
+	u32_t rem_ms, rem_ticks, exp_ticks;
+	u64_t now;
+
+	k_usleep(1); /* align to tick */
 
 	init_timer_data();
 	k_timer_start(&remain_timer, K_MSEC(DURATION), K_NO_WAIT);
 	busy_wait_ms(DURATION / 2);
-	remaining = k_timer_remaining_get(&remain_timer);
+	now = k_uptime_ticks();
+	rem_ms = k_timer_remaining_get(&remain_timer);
+	rem_ticks = k_timer_remaining_ticks(&remain_timer);
+	exp_ticks = k_timer_expires_ticks(&remain_timer);
 	k_timer_stop(&remain_timer);
 
 	/*
@@ -534,7 +541,47 @@ void test_timer_remaining_get(void)
 	 * the value obtained through k_timer_remaining_get() could be larger
 	 * than actual remaining time with maximum error equal to one tick.
 	 */
-	zassert_true(remaining <= (DURATION / 2) + k_ticks_to_ms_floor64(1), NULL);
+	zassert_true(rem_ms <= (DURATION / 2) + k_ticks_to_ms_floor64(1),
+		     NULL);
+
+	zassert_true(rem_ticks <= dur_ticks / 2, NULL);
+	zassert_true((exp_ticks - now) <= dur_ticks / 2, NULL);
+}
+
+void test_timeout_abs(void)
+{
+#ifdef CONFIG_TIMEOUT_64BIT
+	const int expiration = 10000000; /* 10M ticks */
+	k_timeout_t t = K_TIMEOUT_ABS_TICKS(10000000), t2;
+
+	/* Check the other generator macros to make sure they produce
+	 * the same (whiteboxed) converted values
+	 */
+	t2 = K_TIMEOUT_ABS_MS(k_ticks_to_ms_ceil64(expiration));
+	zassert_true(t2.ticks == t.ticks, NULL);
+
+	t2 = K_TIMEOUT_ABS_US(k_ticks_to_us_ceil64(expiration));
+	zassert_true(t2.ticks == t.ticks, NULL);
+
+	t2 = K_TIMEOUT_ABS_NS(k_ticks_to_ns_ceil64(expiration));
+	zassert_true(t2.ticks == t.ticks, NULL);
+
+	t2 = K_TIMEOUT_ABS_CYC(k_ticks_to_cyc_ceil64(expiration));
+	zassert_true(t2.ticks == t.ticks, NULL);
+
+	/* Now set the timeout and make sure the expiration time is
+	 * correct vs. current time.  Tick units and tick alignment
+	 * makes this math exact: remember to add one to match the
+	 * convention (i.e. a timer of "1 tick" will expire at "now
+	 * plus 2 ticks", because "now plus one" will always be
+	 * somewhat less than a tick).
+	 */
+	k_usleep(1); /* align to tick */
+	k_timer_start(&remain_timer, t, K_FOREVER);
+	zassert_true(k_timer_remaining_ticks(&remain_timer)
+		     + k_uptime_ticks() + 1 == expiration, NULL);
+	k_timer_stop(&remain_timer);
+#endif
 }
 
 static void timer_init(struct k_timer *timer, k_timer_expiry_t expiry_fn,
@@ -570,6 +617,7 @@ void test_main(void)
 			 ztest_user_unit_test(test_timer_status_sync),
 			 ztest_user_unit_test(test_timer_k_define),
 			 ztest_user_unit_test(test_timer_user_data),
-			 ztest_user_unit_test(test_timer_remaining_get));
+			 ztest_user_unit_test(test_timer_remaining),
+			 ztest_user_unit_test(test_timeout_abs));
 	ztest_run_test_suite(timer_api);
 }