qcacmn: Wait for scheduler buffers before we panic

The scheduler can run out of buffers and we panic
whenever this happens. However in the older code
we wait until the failure count of getting scheduler
buffers reaches a defined limit, thus making buffers
available in due time and the panic was seen in rare cases,
where failure count was exceeding the defined limit.

With the qdf_flex_mem supprt added for scheduler, the
wait to panic, when the failure count was removed and
we panic when the first failure to get scheduler buffer
is reached. This change adds to wait until the failure
count reaches the limit and then panic.

Change-Id: Ie8774830dc0cefdfccb1473216a9300609ab65bb
CRs-Fixed: 2322049

scheduler/src/scheduler_core.c

@@ -116,6 +116,7 @@ static void scheduler_mq_deinit(struct scheduler_mq_type *msg_q)
 }
 
 static qdf_atomic_t __sched_queue_depth;
+static qdf_atomic_t __sched_dup_fail_count;
 
 static QDF_STATUS scheduler_all_queues_init(struct scheduler_ctx *sched_ctx)
 {
@@ -247,10 +248,15 @@ struct scheduler_msg *scheduler_core_msg_dup(struct scheduler_msg *msg)
 
 	qdf_mem_copy(dup, msg, sizeof(*dup));
 
+	qdf_atomic_set(&__sched_dup_fail_count, 0);
+
 	return dup;
 
 buffer_full:
-	QDF_DEBUG_PANIC("Scheduler buffer is full");
+	if (qdf_atomic_inc_return(&__sched_dup_fail_count) >
+	    SCHEDULER_WRAPPER_MAX_FAIL_COUNT)
+		QDF_DEBUG_PANIC("Scheduler buffer is full");
+
 
 dec_queue_count:
 	qdf_atomic_dec(&__sched_queue_depth);