bug in sf_exp.c in mathfp of newlib

[Ran Cabell]

sf_exp.c in mathfp has a typo - DOUBLE_IS_32BITS is written
DOUBLE_IS_32_BITS (one extra underscore) - the following patch fixes it.

diff -uNr -x *~ newlib-1.8.2.vanilla/newlib/libm/mathfp/sf_exp.c
newlib-1.8.2/newlib/libm/mathfp/sf_exp.c
--- newlib-1.8.2.vanilla/newlib/libm/mathfp/sf_exp.c	Tue Nov 10 21:57:35
1998
+++ newlib-1.8.2/newlib/libm/mathfp/sf_exp.c	Mon Feb 21 14:24:54 2000
@@ -82,11 +82,11 @@
   return (ldexpf (R, N));
 }
 
-#ifdef _DOUBLE_IS_32_BITS
+#ifdef _DOUBLE_IS_32BITS
 
 double exp (double x)
 {
   return (double) expf ((float) x);
 }
 
-#endif /* _DOUBLE_IS_32_BITS */
+#endif /* _DOUBLE_IS_32BITS */


Ran Cabell
rcabell@norfolk.infi.net