Re: Bootstrapping CVS HEAD fails without a Fortran compiler.

[Benoit Sigoure]

Quoting Ralf Wildenhues <address@hidden>:

> Hello Benoit,
>
> * Benoit Sigoure wrote on Wed, Apr 04, 2007 at 12:42:52AM CEST:
> > [...]
> > checking whether we are using the GNU Fortran 77 compiler... no
> > checking whether  accepts -g... no
> > checking how to get verbose linking output from ... configure: WARNING:
> > compilation failed
> > checking for Fortran 77 libraries of ...
> > checking for dummy main to link with Fortran 77 libraries... none
> > checking for Fortran 77 name-mangling scheme... configure: error: 
> > cannot compile
> > a simple Fortran program
> > See `config.log' for more details.
> > make: *** [tests/f77demo/Makefile] Error 1
>
> Please try
>  ./configure F77=no FC=no
>
> > I've never seen this failure before and I never had a Fortran 
> > compiler installed
> > although I already bootstrapped and used CVS libtool.  Is it me 
> > or... is this
> > intended somehow?
>
> Bug in Autoconf and/or Libtool (or their interaction).

Is there a bug report somewhere that explains this issue?  I did not find
anything on savannah...  Are you keeping bug reports / open issues on another
site?

> > BTW, I can simply run ./configure && make check but distcheck still 
> > fails in the
> > same way.  It also fails if I do a VPATH build.
>
> I guess for distcheck you'd need
>  make distcheck DISTCHECK_CONFIGURE_FLAGS='F77=no FC=no'
>
> then (untested).

It still fails the same way.

I'm just starting to remember that I usually use CVS autoconf/automake 
with CVS
libtool, and I get this problem when I use MacPorts' autoconf 2.61 / automake
1.10 / libtool 1.5.22...

*tries to bootstrap with the last CVS snapshot I compiled on my system 
two weeks
ago*

Nope it also fails in the same way with my last CVS autoconf 2.61b / automake
1.10a / GNU libtool 1.2432 2007/03/09 15:08:43 2.1a

Thanks.

--
Benoit Sigoure aka Tsuna
EPITA Research and Development Laboratory