DP: provide data to next LL module no earlier than DP deadline

lets assume DP with 10ms period (a.k.a a deadline).
It starts and finishes earlier, i.e. in 2ms providing 10ms of data
LL starts consuming data in 1ms chunks and will drain
10ms buffer in 10ms, expecting a new portion of data on 11th ms

BUT - the DP module deadline is still 10ms,
regardless if it had finished earlier and it is completely fine
that processing in next cycle takes full 10ms - as long as it
fits into the deadline.

It may lead to underruns:

LL1 (1ms) ---> DP (10ms) -->LL2 (1ms)

ticks 0..9 -> LL1 is producing 1ms data portions,
             DP is waiting, LL2 is waiting
tick 10 - DP has enough data to run, it starts processing
tick 12 - DP finishes earlier, LL2 starts consuming,
          LL1 is producing data
ticks 13-19 LL1 is producing data,
            LL2 is consuming data (both in 1ms chunks)
tick 20  - DP starts processing a new portion of 10ms data,
           having 10ms to finish
	      !!!! but LL2 has already consumed 8ms !!!!
tick 22 - LL2 is consuming the last 1ms data chunk
tick 23 - DP is still processing, LL2 has no data to process
	 			!!! UNDERRUN !!!!
tick 19 - DP finishes properly in a deadline time

Solution: even if DP finishes before its deadline,
the data must be held till deadline time, so LL2 may
start processing no earlier than tick 20

Signed-off-by: Marcin Szkudlinski <marcin.szkudlinski@intel.com>

src/audio/module_adapter/module_adapter.c

@@ -1071,6 +1071,9 @@ static int module_adapter_copy_dp_queues(struct comp_dev *dev)
 		dp_queue = dp_queue_get_next_item(dp_queue);
 	}
 
+	if (mod->dp_startup_delay)
+		return 0;
+
 	dp_queue = dp_queue_get_first_item(&mod->dp_queue_dp_to_ll_list);
 	list_for_item(blist, &dev->bsink_list) {
 		/* output - we need to copy data from dp_queue (as source)

src/include/sof/audio/module_adapter/module/generic.h

@@ -225,6 +225,35 @@ struct processing_module {
 	/* module-specific flags for comp_verify_params() */
 	uint32_t verify_params_flags;
 
+	/* indicates that this DP module did not yet reach its first deadline and
+	 * no data should be passed yet to next LL module
+	 *
+	 * why: lets assume DP with 10ms period (a.k.a a deadline). It starts and finishes
+	 * Earlier, i.e. in 2ms providing 10ms of data. LL starts consuming data in 1ms chunks and
+	 * will drain 10ms buffer in 10ms, expecting a new portion of data on 11th ms
+	 * BUT - the DP module deadline is still 10ms, regardless if it had finished earlier
+	 * and it is completely fine that processing in next cycle takes full 10ms - as long as it
+	 * fits into the deadline.
+	 * It may lead to underruns:
+	 *
+	 * LL1 (1ms) ---> DP (10ms) -->LL2 (1ms)
+	 *
+	 * ticks 0..9 -> LL1 is producing 1ms data portions, DP is waiting, LL2 is waiting
+	 * tick 10 - DP has enough data to run, it starts processing
+	 * tick 12 - DP finishes earlier, LL2 starts consuming, LL1 is producing data
+	 * ticks 13-19 LL1 is producing data, LL2 is consuming data (both in 1ms chunks)
+	 * tick 20  - DP starts processing a new portion of 10ms data, having 10ms to finish
+	 *	      !!!! but LL2 has already consumed 8ms !!!!
+	 * tick 22 - LL2 is consuming the last 1ms data chunk
+	 * tick 23 - DP is still processing, LL2 has no data to process
+	 *			!!! UNDERRUN !!!!
+	 * tick 19 - DP finishes properly in a deadline time
+	 *
+	 * Solution: even if DP finishes before its deadline, the data must be held till
+	 * deadline time, so LL2 may start processing no earlier than tick 20
+	 */
+	bool dp_startup_delay;
+
 	/* flag to indicate module does not pause */
 	bool no_pause;
 

src/schedule/zephyr_dp_schedule.c

@@ -35,10 +35,12 @@ struct scheduler_dp_data {
 
 struct task_dp_pdata {
 	k_tid_t thread_id;		/* zephyr thread ID */
-	uint32_t period_clock_ticks;	/* period the task should be scheduled in Zephyr ticks */
+	uint32_t deadline_clock_ticks;	/* dp module deadline in Zephyr ticks */
+	uint32_t deadline_ll_cycles;	/* dp module deadline in LL cycles */
 	k_thread_stack_t __sparse_cache *p_stack;	/* pointer to thread stack */
 	struct k_sem sem;		/* semaphore for task scheduling */
 	struct processing_module *mod;	/* the module to be scheduled */
+	uint32_t ll_cycles_to_deadline;  /* current number of LL cycles till deadline */
 };
 
 /* Single CPU-wide lock
@@ -227,11 +229,20 @@ void scheduler_dp_ll_tick(void *receiver_data, enum notify_id event_type, void *
 	lock_key = scheduler_dp_lock();
 	list_for_item(tlist, &dp_sch->tasks) {
 		curr_task = container_of(tlist, struct task, list);
+		pdata = curr_task->priv_data;
+		struct processing_module *mod = pdata->mod;
+
+		/* decrease number of LL ticks/cycles left till the module reaches its deadline */
+		if (pdata->ll_cycles_to_deadline) {
+			pdata->ll_cycles_to_deadline--;
+			if (!pdata->ll_cycles_to_deadline)
+				/* deadline reached, clear startup delay flag.
+				 * see dp_startup_delay comment for details
+				 */
+				mod->dp_startup_delay = false;
+		}
 
-		/* step 1 - check if the module is ready for processing */
 		if (curr_task->state == SOF_TASK_STATE_QUEUED) {
-			pdata = curr_task->priv_data;
-			struct processing_module *mod = pdata->mod;
 			bool mod_ready;
 
 			mod_ready = module_is_ready_to_process(mod, mod->sources,
@@ -240,7 +251,9 @@ void scheduler_dp_ll_tick(void *receiver_data, enum notify_id event_type, void *
 							       mod->num_of_sinks);
 			if (mod_ready) {
 				/* set a deadline for given num of ticks, starting now */
-				k_thread_deadline_set(pdata->thread_id, pdata->period_clock_ticks);
+				k_thread_deadline_set(pdata->thread_id,
+						      pdata->deadline_clock_ticks);
+				pdata->ll_cycles_to_deadline = pdata->deadline_ll_cycles;
 
 				/* trigger the task */
 				curr_task->state = SOF_TASK_STATE_RUNNING;
@@ -352,7 +365,7 @@ static int scheduler_dp_task_shedule(void *data, struct task *task, uint64_t sta
 	struct scheduler_dp_data *dp_sch = (struct scheduler_dp_data *)data;
 	struct task_dp_pdata *pdata = task->priv_data;
 	unsigned int lock_key;
-	uint64_t period_clock_ticks;
+	uint64_t deadline_clock_ticks;
 
 	lock_key = scheduler_dp_lock();
 
@@ -371,13 +384,16 @@ static int scheduler_dp_task_shedule(void *data, struct task *task, uint64_t sta
 	task->state = SOF_TASK_STATE_QUEUED;
 	list_item_prepend(&task->list, &dp_sch->tasks);
 
-	period_clock_ticks = period * CONFIG_SYS_CLOCK_TICKS_PER_SEC;
-	/* period is in us - convert to seconds in next step
-	 * or it always will be zero because of fixed point calculation
+	deadline_clock_ticks = period * CONFIG_SYS_CLOCK_TICKS_PER_SEC;
+	/* period/deadline is in us - convert to seconds in next step
+	 * or it always will be zero because of integer calculation
 	 */
-	period_clock_ticks /= 1000000;
+	deadline_clock_ticks /= 1000000;
 
-	pdata->period_clock_ticks = period_clock_ticks;
+	pdata->deadline_clock_ticks = deadline_clock_ticks;
+	pdata->deadline_ll_cycles = period / LL_TIMER_PERIOD_US;
+	pdata->ll_cycles_to_deadline = 0;
+	pdata->mod->dp_startup_delay = true;
 	scheduler_dp_unlock(lock_key);
 
 	tr_dbg(&dp_tr, "DP task scheduled with period %u [us]", (uint32_t)period);