correct M_MAX_D for old floating point format

diff --git a/include/_tail_cc/math.h b/include/_tail_cc/math.h
index 3265905..705743a 100644 (file)
--- a/include/_tail_cc/math.h
+++ b/include/_tail_cc/math.h
@@ -63,7 +63,7 @@ extern double sinh(), tanh(), cosh();
 #endif
 
 #ifndef M_MAX_D
-#define M_MAX_D                 8.988465674311561576e+307
+#define M_MAX_D          4.494232837155780788e+307
 #define M_MIN_D                 5.562684646268003458e-309
 #define M_DEXPLEN       11
 #endif