Implemented GetStartLatency

git-svn-id: file:///srv/svn/repos/haiku/trunk/current@5626 a95241bf-73f2-0310-859d-f6bbb57e9c96

src/kits/media/MediaRoster.cpp

@@ -2422,8 +2422,25 @@ status_t
 BMediaRoster::GetStartLatencyFor(const media_node & time_source,
 								 bigtime_t * out_latency)
 {
-	UNIMPLEMENTED();
+	CALLED();
-	*out_latency = 5000;
+	if (out_latency == NULL)
+		return B_BAD_VALUE;
+	if (IS_INVALID_NODE(time_source))
+		return B_MEDIA_BAD_NODE;
+	if ((time_source.kind & B_TIME_SOURCE) == 0)
+		return B_MEDIA_BAD_NODE;
+	
+	timesource_get_start_latency_request request;
+	timesource_get_start_latency_reply reply;
+	status_t rv;
+
+	rv = QueryPort(time_source.port, TIMESOURCE_GET_START_LATENCY, &request, sizeof(request), &reply, sizeof(reply));
+	if (rv != B_OK)
+		return rv;
+	
+	*out_latency = reply.start_latency;
+	
+	printf("BMediaRoster::GetStartLatencyFor timesource %ld has maximum initial latency %Ld\n", time_source.node, *out_latency);
 	return B_OK;
 }
 

src/kits/media/TimeSource.cpp

@@ -235,7 +235,7 @@ BTimeSource::HandleMessage(int32 message,
 						   size_t size)
 {
 	PRINT(4, "BTimeSource::HandleMessage %#lx, node %ld\n", message, fNodeID);
-
+	status_t rv;
 	switch (message) {
 		case TIMESOURCE_OP:
 		{
@@ -275,6 +275,15 @@ BTimeSource::HandleMessage(int32 message,
 			DirectRemoveMe(data->node);
 			return B_OK;
 		}
+		
+		case TIMESOURCE_GET_START_LATENCY:
+		{
+			const timesource_get_start_latency_request *request = static_cast<const timesource_get_start_latency_request *>(rawdata);
+			timesource_get_start_latency_reply reply;
+			rv = GetStartLatency(&reply.start_latency);
+			request->SendReply(rv, &reply, sizeof(reply));
+			return B_OK;
+		}
 	}
 	return B_ERROR;
 }