Re: autoconf-2.59b fails to include correct X11R6 paths in configure scripts on AMD64 systems

[Pavel Roskin]

Hello again,

On Tue, 2005-09-06 at 12:29 -0700, Paul Eggert wrote:
> In <http://lists.gnu.org/archive/html/bug-autoconf/2005-07/msg00052.html>
> "Nelson H. F. Beebe" <address@hidden> writes:
> 
> > The latest release of autoconf-2.59b, dated 28-Dec-2005, fails to
> > include correct X11R6 paths in configure scripts on AMD64 systems
> > running GNU/Linux.
> 
> Thanks for reporting this.  I have installed the XMKMF patch suggested
> by Pavel Roskin in
> <http://lists.gnu.org/archive/html/autoconf-patches/2005-08/msg00078.html>.
> Does this fix the problem?  (Sorry, I'm not an expert on these systems
> yet, so I have to ask dumb questions like this.  :-)

I didn't realize the question was about 64-bit compilation.  In this
case, xmkmf should have found X11.  It actually finds it on Fedora Core
4.  The fallback to _AC_PATH_X_DIRECT should not be needed on any modern
systems.

In my opinion, we should not add /lib64 paths to _AC_PATH_X_DIRECT.  My
arguments are:

1) It a misconfiguration of free software that can be easily fixed. 
Fedora did it right, so others can fix it too.  It may even be a local
misconfiguration.

2) It's easier than fixing imake to sense $CC and not to hardcode the
architecture flags.  It's just making imake work the way it does for the
native architecture.

3) We can see proliferation of lib* in the future.  It's not like there
is just one 64-bit programming model (LP64) and one 64-bit instruction
set.  We may see migration from LP64 to ILP64 some day, and then it
won't be lib vs lib64, but rather lib64 vs e.g. libi64.

-- 
Regards,
Pavel Roskin