This is the mail archive of the libstdc++@gcc.gnu.org mailing list for the libstdc++ project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]

[libstdc++ 3.0] GLIBCPP_ENABLE_LONG_LONG fix with AC_LANG backported

[Get raw message]
A week ago I committed a fix for 'long long' checking to trunk and branch.
Another fix was needed later and applied to the trunk earlier this week;
here is the same change for the branch.  Tested on i686/linux with no
problems, but the earlier fix was for hppa.  I'll be keeping an eye out
for hppa weirdness on the branch.


2001-11-09  Phil Edwards  <pme@gcc.gnu.org>

	* acinclude.m4 (GLIBCPP_ENABLE_LONG_LONG):  Run the test in
	LANG_CPLUSPLUS mode.
	* aclocal.m4:  Regenerate.
	* configure:  Regenerate.

Index: acinclude.m4
===================================================================
RCS file: /cvs/gcc/gcc/libstdc++-v3/acinclude.m4,v
retrieving revision 1.129.2.30
diff -u -3 -p -r1.129.2.30 acinclude.m4
--- acinclude.m4	2001/11/02 22:39:01	1.129.2.30
+++ acinclude.m4	2001/11/09 21:33:03
@@ -1378,8 +1378,10 @@ AC_DEFUN(GLIBCPP_ENABLE_LONG_LONG, [dnl
    esac],
   enable_long_long=GLIBCPP_ENABLE_LONG_LONG_DEFAULT)dnl
 
-  AC_MSG_CHECKING([for enabled long long I/O support])
+  AC_LANG_SAVE
+  AC_LANG_CPLUSPLUS
 
+  AC_MSG_CHECKING([for enabled long long I/O support])
   # iostreams require strtoll, strtoull to compile
   AC_TRY_COMPILE([#include <stdlib.h>],
                  [char* tmp; strtoll("gnu", &tmp, 10);],,[enable_long_long=no])
@@ -1391,6 +1393,8 @@ AC_DEFUN(GLIBCPP_ENABLE_LONG_LONG, [dnl
     AC_DEFINE(_GLIBCPP_USE_LONG_LONG)
   fi
   AC_MSG_RESULT($enable_long_long)
+
+  AC_LANG_RESTORE
 ])
 
 


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]