m4-patches
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: FYI: 39-gary-fix-gmp-test.patch reverted


From: Akim Demaille
Subject: Re: FYI: 39-gary-fix-gmp-test.patch reverted
Date: Fri, 29 Aug 2003 12:14:44 +0200
User-agent: Gnus/5.1002 (Gnus v5.10.2) Emacs/21.3 (gnu/linux)

 > Salut Akim!
 >> This patch is wrong, since the goal of ./testsuite is to be able to be
 >> run easily by hand.  The functionality of TESTS_ENVIRONMENT you are
 >> looking for is in atlocal.  If GMP does not work, dig in there.

 > Ah yes.  At some point atlocal stopped being generated.  I am
 > reverting this patch in favour of adding `tests/atlocal' to
 > AC_CONFIG_FILES.  Is that the right way to do things?

Something is weird in here.  This should be done by
AC_CONFIG_TESTDIR([tests]).  That's what the doc says, but the code no
longer does this.  Autoconf and Bison include their own
AC_CONFIG_FILES for atlocal.  I don't know why, but I'm in favor of
restoring the documented behavior.  But I guess that in the meanwhile
M4 ought to have its AC_CONFIG_TESTDIR([tests/atlocal]), indeed.

 >> I still cannot build M4, so I can't help.  Libtool is definitely not
 >> helping me figuring out what's wrong :(

 > Send me a termlog and I'll see if I can help:

Finally it works!  I had to change one thing in CVS Autoconf.

 > Maybe it is just your version of libtool is not new enough?

A big problem I had was that libtool accepts severely mixed setups,
with some files of some version of libtool, and others from another
version.  But since they don't check each other, it just fails with no
clear sign.

This time, it works fine.  I think that the patch from Alexandre
solved the issue.

There remains one warning with CVS Automake:

autoreconf: running: automake --add-missing --copy --force-missing
configure.ac:194: required file `./${top_srcdir}/m4/gnu-obstack.h' not found
autoreconf: automake failed with exit status: 1




reply via email to

[Prev in Thread] Current Thread [Next in Thread]