diff --git a/include/kernel.h b/include/kernel.h
index 419a1b84106..6db33e6bb61 100644
--- a/include/kernel.h
+++ b/include/kernel.h
@@ -466,6 +466,10 @@ struct _thread_base {
 		u16_t preempt;
 	};
 
+#ifdef CONFIG_SCHED_DEADLINE
+	int prio_deadline;
+#endif
+
 	u32_t order_key;
 
 #ifdef CONFIG_SMP
@@ -1000,6 +1004,38 @@ __syscall int k_thread_priority_get(k_tid_t thread);
  */
 __syscall void k_thread_priority_set(k_tid_t thread, int prio);
 
+
+#ifdef CONFIG_SCHED_DEADLINE
+/**
+ * @brief Set deadline expiration time for scheduler
+ *
+ * This sets the "deadline" expiration as a time delta from the
+ * current time, in the same units used by k_cycle_get_32().  The
+ * scheduler (when deadline scheduling is enabled) will choose the
+ * next expiring thread when selecting between threads at the same
+ * static priority.  Threads at different priorities will be scheduled
+ * according to their static priority.
+ *
+ * @note Deadlines that are negative (i.e. in the past) are still seen
+ * as higher priority than others, even if the thread has "finished"
+ * its work.  If you don't want it scheduled anymore, you have to
+ * reset the deadline into the future, block/pend the thread, or
+ * modify its priority with k_thread_priority_set().
+ *
+ * @note Despite the API naming, the scheduler makes no guarantees the
+ * the thread WILL be scheduled within that deadline, nor does it take
+ * extra metadata (like e.g. the "runtime" and "period" parameters in
+ * Linux sched_setattr()) that allows the kernel to validate the
+ * scheduling for achievability.  Such features could be implemented
+ * above this call, which is simply input to the priority selection
+ * logic.
+ *
+ * @param thread A thread on which to set the deadline
+ * @param deadline A time delta, in cycle units
+ */
+__syscall void k_thread_deadline_set(k_tid_t thread, int deadline);
+#endif
+
 /**
  * @brief Suspend a thread.
  *
diff --git a/kernel/Kconfig b/kernel/Kconfig
index 4694c8bb636..18ff7af6e83 100644
--- a/kernel/Kconfig
+++ b/kernel/Kconfig
@@ -122,6 +122,18 @@ config NUM_METAIRQ_PRIORITIES
 	 for cooperative threads, this tool probably shouldn't be used
 	 from application code.
 
+config SCHED_DEADLINE
+	bool
+	prompt "Enable earliest-deadline-first scheduling"
+	default n
+	help
+	  This enables a simple "earliest deadline first" scheduling
+	  mode where threads can set "deadline" deltas measured in
+	  k_cycle_get_32() units.  Priority decisions within (!!) a
+	  single priority will choose the next expiring deadline and
+	  not simply the least recently added thread.
+
+
 config MAIN_STACK_SIZE
 	int
 	prompt "Size of stack for initialization and main thread"
diff --git a/kernel/include/ksched.h b/kernel/include/ksched.h
index 65d280450f2..4cfc3af722f 100644
--- a/kernel/include/ksched.h
+++ b/kernel/include/ksched.h
@@ -213,11 +213,7 @@ static inline int _is_prio_lower_or_equal(int prio1, int prio2)
 	return _is_prio1_lower_than_or_equal_to_prio2(prio1, prio2);
 }
 
-static inline int _is_t1_higher_prio_than_t2(struct k_thread *t1,
-					     struct k_thread *t2)
-{
-	return _is_prio1_higher_than_prio2(t1->base.prio, t2->base.prio);
-}
+int _is_t1_higher_prio_than_t2(struct k_thread *t1, struct k_thread *t2);
 
 static inline int _is_valid_prio(int prio, void *entry_point)
 {
diff --git a/kernel/sched.c b/kernel/sched.c
index 729716e1a8c..c7fa3bb3884 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -87,6 +87,31 @@ s32_t _ms_to_ticks(s32_t ms)
 }
 #endif
 
+int _is_t1_higher_prio_than_t2(struct k_thread *t1, struct k_thread *t2)
+{
+	if (t1->base.prio < t2->base.prio) {
+		return 1;
+	}
+
+#ifdef CONFIG_SCHED_DEADLINE
+	/* Note that we don't care about wraparound conditions.  The
+	 * expectation is that the application will have arranged to
+	 * block the threads, change their priorities or reset their
+	 * deadlines when the job is complete.  Letting the deadlines
+	 * go negative is fine and in fact prevents aliasing bugs.
+	 */
+	if (t1->base.prio == t2->base.prio) {
+		int now = (int) k_cycle_get_32();
+		int dt1 = t1->base.prio_deadline - now;
+		int dt2 = t2->base.prio_deadline - now;
+
+		return dt1 < dt2;
+	}
+#endif
+
+	return 0;
+}
+
 static struct k_thread *next_up(void)
 {
 #ifndef CONFIG_SMP
@@ -607,6 +632,36 @@ Z_SYSCALL_HANDLER(k_thread_priority_set, thread_p, prio)
 }
 #endif
 
+#ifdef CONFIG_SCHED_DEADLINE
+void _impl_k_thread_deadline_set(k_tid_t tid, int deadline)
+{
+	struct k_thread *th = tid;
+
+	LOCKED(&sched_lock) {
+		th->base.prio_deadline = k_cycle_get_32() + deadline;
+		if (_is_thread_queued(th)) {
+			_priq_run_remove(&_kernel.ready_q.runq, th);
+			_priq_run_add(&_kernel.ready_q.runq, th);
+		}
+	}
+}
+
+#ifdef CONFIG_USERSPACE
+Z_SYSCALL_HANDLER(k_thread_deadline_set, thread_p, deadline)
+{
+	struct k_thread *thread = (struct k_thread *)thread_p;
+
+	Z_OOPS(Z_SYSCALL_OBJ(thread, K_OBJ_THREAD));
+	Z_OOPS(Z_SYSCALL_VERIFY_MSG(deadline > 0,
+				    "invalid thread deadline %d",
+				    (int)deadline));
+
+	_impl_k_thread_deadline_set((k_tid_t)thread, deadline);
+	return 0;
+}
+#endif
+#endif
+
 void _impl_k_yield(void)
 {
 	__ASSERT(!_is_in_isr(), "");
diff --git a/kernel/thread.c b/kernel/thread.c
index cdc074b4315..827a984fa51 100644
--- a/kernel/thread.c
+++ b/kernel/thread.c
@@ -313,6 +313,9 @@ void _setup_new_thread(struct k_thread *new_thread,
 	if (options & K_INHERIT_PERMS) {
 		_thread_perms_inherit(_current, new_thread);
 	}
+#endif
+#ifdef CONFIG_SCHED_DEADLINE
+	new_thread->base.prio_deadline = 0;
 #endif
 	new_thread->resource_pool = _current->resource_pool;
 }