Re: ultrix and 'missing'?

[Harlan Stenn]

The failure is happening on a SunOS 4.1.3 box with "cc" and GNU make
(3.76.1).  The build is being done outside of the source tree:

 % mkdir A.`config.guess`
 % cd !$
 % ../configure
 % make

As I recall, the problem did not happen on a 4.1.3 box with gcc and GNU
make 3.74 .

I'm trying to install CVS automake now (from subversions.gnu.org - is
this copy different form the one at sourceware.cygnus.com?).  I have to
upgrade perl first (5.003 is installed at the moment).

OK, this won't happen with the stock C compiler.  This will take some
work.  I'll see what I can do.  I'll need to do this on some other box
and then try the tests on the 4.1.3/cc box.

When will ansi5.test be checked in?

H
--
> What I'd really like is a way to see the failure with GNU make and gcc
> (ansi5.test does this and then does a couple small hacks to fool the
> build into thinking that ansi2knr is required).  If I had this then I
> could debug it here.
> 
> Failing that, a simple test which fails on your box with your compiler
> would help.  I might have access to a similar box, in which case I
> could find out what is wrong.
> 
> One problem is that right now I don't have a theory of what is going
> wrong.  I looked at the Makefiles from ansi5.test and they look fine
> to me.  And if I can't reproduce it, unfortunately that means you'll
> probably have to figure out how to debug it :-(.
> 
> What system are you using?  What make are you using?  What compiler?
> Are you building the source tree or in a separate tree?
> 
> Sometimes `make' will have a flag that prints debugging information.
> So you might take a look through that to see if it is interesting.
> Perhaps we've simply run into a bug or limitation in your make which
> we must then work around.
> 
> Tom