Re: CVS Build problem (what.m missing?)

[John Swensen]

Fredrik Lingvall wrote:
> John W. Eaton wrote:
> > Earlier, you showed a .f file that started with
> >
> > program foo
> >       double complex ...
> >
> > If that is what the above shell script produces, then I don't know how
> > to fix this problem, as your shell is doing something completely
> > unexpected.  So, is the code above copied into the conftest.f file
> > correctly or is your shell broken?
> >   
> No it is not what the shell script produces, it was just a 
> copy-and-paste error
> (and I didn't know that Fortran was sensitive to white space). I don't 
> think
> that this is what causing the problem anyway. Perhaps the problem goes 
> away in a future
> Gentoo update (since it works for x86_64 branch). For know I just 
> temporary remove the test code
> from the configure script and Octave builds fine.
>
> > | XLIBS="$LIBS"
> > | LIBS="$BLAS_LIBS $FLIBS $LIBS"
> > | ac_ext=f
> > | ac_compile='$F77 -c $FFLAGS conftest.$ac_ext >&5'
> > | ac_link='$F77 -o conftest$ac_exeext $FFLAGS $LDFLAGS conftest.$ac_ext
> > | $LIBS >&5'
> > | ac_compiler_gnu=$ac_cv_f77_compiler_gnu
> > | | (eval "$ac_compile"; eval "$ac_link") 2>&5
> > | ac_ext=c
> > | ac_cpp='$CPP $CPPFLAGS'
> > | ac_compile='$CC -c $CFLAGS $CPPFLAGS conftest.$ac_ext >&5'
> > | ac_link='$CC -o conftest$ac_exeext $CFLAGS $CPPFLAGS $LDFLAGS
> > | conftest.$ac_ext $LIBS >&5'
> > | ac_compiler_gnu=$ac_cv_c_compiler_gnu
> > | | LIBS="$XLIBS"
> > | case "`./conftest$ac_exeext`" in
> > |   *succeeded*)
> > |   { echo "$as_me:$LINENO: result: yes" >&5
> > | echo "${ECHO_T}yes" >&6; }
> > | ;;
> > |   *)
> > |   { echo "$as_me:$LINENO: result: no" >&5
> > | echo "${ECHO_T}no" >&6; }
> > |   { echo "$as_me:$LINENO: WARNING: Your BLAS library was apparently
> > | compiled with a Fortran" >&5
> > | echo "$as_me: WARNING: Your BLAS library was apparently compiled 
> > with a
> > | Fortran" >&2;}
> > |   { echo "$as_me:$LINENO: WARNING: compiler that uses a different
> > | calling convention from" >&5
> > | echo "$as_me: WARNING: compiler that uses a different calling 
> > convention
> > | from" >&2;}
> > |   { echo "$as_me:$LINENO: WARNING: the one used by the selected
> > | compiler, $F77." >&5
> > | echo "$as_me: WARNING: the one used by the selected compiler, 
> > $F77." >&2;}
> > |   { { echo "$as_me:$LINENO: error: You must correct this problem 
> > before
> > | building Octave." >&5
> > | echo "$as_me: error: You must correct this problem before building
> > | Octave." >&2;}
> > |    { (exit 1); exit 1; }; }
> > | ;;
> > | esac
> > | | | If I have understood this correctly then the lines:
> > | | ac_ext=f
> > | ac_compile='$F77 -c $FFLAGS conftest.$ac_ext >&5'
> > | ac_link='$F77 -o conftest$ac_exeext $FFLAGS $LDFLAGS conftest.$ac_ext
> > | $LIBS >&5'
> > | ac_compiler_gnu=$ac_cv_f77_compiler_gnu
> > | | (eval "$ac_compile"; eval "$ac_link") 2>&5
> > | | is for compiling the Fortran test code and
> > | | LIBS="$XLIBS"
> > | case "`./conftest$ac_exeext`" in
> > |   *succeeded*)
> > |   { echo "$as_me:$LINENO: result: yes" >&5
> > | echo "${ECHO_T}yes" >&6; }
> > | ;;
> > |   *)
> > | | runs the test code. However, I don't understand what the lines:
> > | | ac_ext=c
> > | ac_cpp='$CPP $CPPFLAGS'
> > | ac_compile='$CC -c $CFLAGS $CPPFLAGS conftest.$ac_ext >&5'
> > | ac_link='$CC -o conftest$ac_exeext $CFLAGS $CPPFLAGS $LDFLAGS
> > | conftest.$ac_ext $LIBS >&5'
> > | ac_compiler_gnu=$ac_cv_c_compiler_gnu
> > | | does. It seems to be for compiling some C-test code that I can't 
> > find.
> >
> > The configure.in file has these lines:
> >
> > AC_LANG_PUSH(Fortran 77)
> > (eval "$ac_compile"; eval "$ac_link") 2>&AS_MESSAGE_LOG_FD
> > AC_LANG_POP(Fortran 77)
> >
> > I think the lines you quote directly above are the result of the
> > AC_LANG_POP macro, which is resetting the autoconf variables to use
> > the C compiler.  The argument to AC_LANG_POP is the *current* language,
> > to allow a consistency check (you should be popping the current
> > language).
> >   
> OK.
>
> /Fredrik
I also have this problem on OSX using the CVS as of 5:18PM EST 
10/31/2007.  My configure line is as follows:

./configure CC=gcc-4 CXX=g++-4 F77=gfortran CPPFLAGS=-I/sw/include 
LDFLAGS=-L/sw/lib --prefix=/sw/opt/octave/cvs_2.9_20071007

The config.log file indicates it failed at:
configure:15915: checking BLAS library calling convention compatibility
configure:15958: result: no

Reading through your email chain, I wasn't sure whether a solution had 
been reached, other than to cut out the test code so it always succeeds, 
but is this going to cause problems when compiling/running?

I tried to compile the conftest code by hand, but if I compile/link it with:
gfortran conftest.f -c -g3
gfortran conftest.o -o conftest  -Wl,-framework -Wl,vecLib
and then try to run it, I get a "Illegal instruction" error.  I ran it 
in gdb and get further information concerning where the error occurred:
Program received signal EXC_BAD_INSTRUCTION, Illegal instruction/operand.
0x8fe12f94 in __dyld_stub_binding_helper_interface ()

John Swensen