advertising a push would make the GPU worker stay awake, and continue trying to push data while that won't work any better unless something actually happens.
@@ -618,8 +618,9 @@ static int __starpu_handle_node_data_requests(struct _starpu_data_request_prio_l
/* Prefetching more there would make the situation even worse */
break;
}
+ else
+ (*pushed)++;
- (*pushed)++;
if (starpu_timing_now() - start >= MAX_PUSH_TIME)
{
/* We have spent a lot of time doing requests, skip pushing more for now */