Re: Bug in LT_PROG_GCJ ?

2007-03-21 Thread Steve Ellcey
 Applied to HEAD.  I put you in THANKS, too.
 
 Cheers,
 Ralf

Thanks Ralf, this is one step towards getting GCC off of its ancient 1.4
libtool.

Steve Ellcey
[EMAIL PROTECTED]


___
Bug-libtool mailing list
Bug-libtool@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-libtool


Re: Bug in LT_PROG_GCJ ?

2007-03-18 Thread Ralf Wildenhues
Hello Steve, and sorry for the delay,

* Ralf Wildenhues wrote on Fri, Mar 16, 2007 at 06:34:34PM CET:
 * Steve Ellcey wrote on Fri, Mar 16, 2007 at 06:30:38PM CET:
AC_DEFUN([LT_PROG_GCJ],
[m4_ifdef([AC_PROG_GCJ], [AC_PROG_GCJ],
  [m4_ifdef([A][M_PROG_GCJ], [A][M_PROG_GCJ],
[AC_CHECK_TOOL(GCJ, gcj,)
  test x${GCJFLAGS+set} = xset || GCJFLAGS=-g -O2
  AC_SUBST(GCJFLAGS)])])dnl
])
 [...]
   Does changing the line to
   AC_SUBST(GCJFLAGS)])])[]dnl
   
   work?

Applied to HEAD.  I put you in THANKS, too.

Cheers,
Ralf

2007-03-18  Ralf Wildenhues  [EMAIL PROTECTED]

* libltdl/m4/libtool.m4 (LT_PROG_GCJ): Avoid M4 expansion error
that caused `dnl' to be merged to the previous word.
* THANKS: Update.
Report by Steve Ellcey.

Index: libltdl/m4/libtool.m4
===
RCS file: /cvsroot/libtool/libtool/libltdl/m4/libtool.m4,v
retrieving revision 1.98
diff -u -r1.98 libtool.m4
--- libltdl/m4/libtool.m4   26 Feb 2007 07:44:23 -  1.98
+++ libltdl/m4/libtool.m4   18 Mar 2007 17:36:03 -
@@ -6844,7 +6844,7 @@
   [m4_ifdef([A][M_PROG_GCJ], [A][M_PROG_GCJ],
 [AC_CHECK_TOOL(GCJ, gcj,)
   test x${GCJFLAGS+set} = xset || GCJFLAGS=-g -O2
-  AC_SUBST(GCJFLAGS)])])dnl
+  AC_SUBST(GCJFLAGS)])])[]dnl
 ])
 
 # Old name:


___
Bug-libtool mailing list
Bug-libtool@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-libtool


Re: Bug in LT_PROG_GCJ ?

2007-03-16 Thread Steve Ellcey
  AC_DEFUN([LT_PROG_GCJ],
  [m4_ifdef([AC_PROG_GCJ], [AC_PROG_GCJ],
[m4_ifdef([A][M_PROG_GCJ], [A][M_PROG_GCJ],
  [AC_CHECK_TOOL(GCJ, gcj,)
test x${GCJFLAGS+set} = xset || GCJFLAGS=-g -O2
AC_SUBST(GCJFLAGS)])])dnl
  ])
  
  I think the 'dnl' at the end of the AC_SUBST line is the problem.  Does
  this seem right?  The problem went away after I removed the 'dnl'.  If
  this looks right can someone check this change into libtool.
 
 Does changing the line to
 AC_SUBST(GCJFLAGS)])])[]dnl
 
 work?
 
 Cheers,
 Ralf

Ralf,

Were you (or maybe someone else) going to make this change in the
libtool sources?  Do I need to submit an official patch?  I don't have
write access to the libtool tree so someone else would have to check it
in anyway.  Your suggested change fixed my problem and I would like to
see it checked into the libtool CVS tree so that I don't have to patch
libtool locally in order to use the latest version.

Steve Ellcey
[EMAIL PROTECTED]


___
Bug-libtool mailing list
Bug-libtool@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-libtool


Re: Bug in LT_PROG_GCJ ?

2007-03-16 Thread Ralf Wildenhues
* Steve Ellcey wrote on Fri, Mar 16, 2007 at 06:30:38PM CET:
   AC_DEFUN([LT_PROG_GCJ],
   [m4_ifdef([AC_PROG_GCJ], [AC_PROG_GCJ],
 [m4_ifdef([A][M_PROG_GCJ], [A][M_PROG_GCJ],
   [AC_CHECK_TOOL(GCJ, gcj,)
 test x${GCJFLAGS+set} = xset || GCJFLAGS=-g -O2
 AC_SUBST(GCJFLAGS)])])dnl
   ])
[...]
  Does changing the line to
  AC_SUBST(GCJFLAGS)])])[]dnl
  
  work?

 Were you (or maybe someone else) going to make this change in the
 libtool sources?

Yes.  However my Libtool work needs to wait until the weekend.  Sorry.

Cheers,
Ralf


___
Bug-libtool mailing list
Bug-libtool@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-libtool


Bug in LT_PROG_GCJ ?

2007-03-13 Thread Steve Ellcey

I am currently expirementing with using the latest (ToT) libtool in the
GCC and binutils builds.  I ran into a problem when trying to run aclocal
and autoconf in the GCC libjava directory, I was getting:

$ aclocal -I . -I .. -I ../config
$ autoconf
configure:15448: error: possibly undefined macro: AM_PROG_GCJdnl
  If this token and others are legitimate, please use m4_pattern_allow.
  See the Autoconf documentation.

I think I tracked this down to the expansion of LT_PROG_GCJ from libtool.m4.

In libltdl/m4/libtool.m4 there is:

# LT_PROG_GCJ
# ---
AC_DEFUN([LT_PROG_GCJ],
[m4_ifdef([AC_PROG_GCJ], [AC_PROG_GCJ],
  [m4_ifdef([A][M_PROG_GCJ], [A][M_PROG_GCJ],
[AC_CHECK_TOOL(GCJ, gcj,)
  test x${GCJFLAGS+set} = xset || GCJFLAGS=-g -O2
  AC_SUBST(GCJFLAGS)])])dnl
])

I think the 'dnl' at the end of the AC_SUBST line is the problem.  Does
this seem right?  The problem went away after I removed the 'dnl'.  If
this looks right can someone check this change into libtool.

Steve Ellcey
[EMAIL PROTECTED]


___
Bug-libtool mailing list
Bug-libtool@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-libtool


Re: Bug in LT_PROG_GCJ ?

2007-03-13 Thread Steve Ellcey
 From: Ralf Wildenhues [EMAIL PROTECTED]
 
 Does changing the line to
 AC_SUBST(GCJFLAGS)])])[]dnl
 
 work?
 
 Cheers,
 Ralf

Yes, that works.

Steve Ellcey
[EMAIL PROTECTED]


___
Bug-libtool mailing list
Bug-libtool@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-libtool