Re: [bug-gnulib] coreutils-6.4 released (stable)

[Bruno Haible]

Matthew,

Paul Eggert wrote:
> Eric Blake <address@hidden> writes:
> 
> > I didn't see this show up on bug-gnulib, nor in the list queue, so I took
> > the liberty of resending it...
> >
> > -------- Original Message --------
> > Date: Wed, 15 Nov 2006 19:13:36 -0600
> > From: Matthew Woehlke <mw_triad AT users.sourceforge.net>
> > ...
> > Ok. Using the latest snapshot Eric provided me with, m4 fails to build
> > on Tandem NSK/OSS. Using the above, *plus* this patch:
> > http://lists.gnu.org/archive/html/bug-gnulib/2006-10/msg00140.html
> > both 'make' and 'make check' succeed.
> 
> As Bruno mentioned in
> <http://lists.gnu.org/archive/html/bug-gnulib/2006-11/msg00092.html>,
> he'd rather not head down this path due to the high cost/benefit
> ratio.  Even if M4 builds with the abovementioned patch, the approach
> will likely cause runtime misbehaviors -- almost certainly in
> coreutils, and quite possibly also in M4.

Yes. Matthew, I suggest that you compile with optimization (CFLAGS="-O" or
something like that). Then, with the patch to m4/longlong.m4 that Paul
committed on 2006-11-07, the configure script should find that 'long long'
is unusable and refrain from using it. You can verify if this is so by
looking whether HAVE_LONG_LONG_INT gets defined in config.h.

Bruno