_Py_normalize_encoding(): explain how the value 6 was computed
diff --git a/Objects/unicodeobject.c b/Objects/unicodeobject.c
index 1375ef3..224a80b 100644
--- a/Objects/unicodeobject.c
+++ b/Objects/unicodeobject.c
@@ -2983,6 +2983,7 @@
     char *l_end;
 
     if (encoding == NULL) {
+        /* 6 == strlen("utf-8") + 1 */
         if (lower_len < 6)
             return 0;
         strcpy(lower, "utf-8");