Device Threads: The default maximum device thread number has been reduced from 1024 to 64, users can still overwrite with the -T option

include/types.h

@@ -1760,7 +1760,7 @@ typedef enum kernel_workload
   KERNEL_LOOPS_MIN   = 1,
   KERNEL_LOOPS_MAX   = 1024,
   KERNEL_THREADS_MIN = 1,
-  KERNEL_THREADS_MAX = 256,
+  KERNEL_THREADS_MAX = 64,
 
 } kernel_workload_t;