Update benchmark: run each bench separetely to avoid bad malloc implementation bug

(Running list benchmark before array fragment the memory too much and
 generate a noticable slow down).

Add Klib bench.
o

bench/bench-stl.cpp

@@ -32,7 +32,7 @@ static void rand_init(void)
 static unsigned int rand_get(void)
 {
   randValue = randValue * 31421U + 6927U;
-  return (int) randValue;
+  return randValue;
 }
 
 static void test_function(const char str[], size_t n, void (*func)(size_t))
@@ -141,9 +141,15 @@ test_dict2(unsigned long  n)
 
 int main(int argc, const char *argv[])
 {
-  test_function("List   time",10000000, test_list);
-  test_function("Array  time", 100000000, test_array);
-  test_function("Rbtree time", 1000000, test_rbtree);
-  test_function("Dict(m)time", 1000000, test_dict1);
-  test_function("Dict(u)time", 1000000, test_dict2);
+  int n = (argc > 1) ? atoi(argv[1]) : 0;
+  if (n == 1)
+    test_function("List   time",10000000, test_list);
+  if (n == 2)
+    test_function("Array  time", 100000000, test_array);
+  if (n == 3)
+    test_function("Rbtree time", 1000000, test_rbtree);
+  if (n == 4)
+    test_function("Dict(m)time", 1000000, test_dict1);
+  if (n == 5)
+    test_function("Dict(u)time", 1000000, test_dict2);
 }