autoconf
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: WARNING: gsl/gsl_sf_bessel.h: accepted by the compiler


From: Dr. David Kirkby
Subject: Re: WARNING: gsl/gsl_sf_bessel.h: accepted by the compiler
Date: Fri, 21 Mar 2003 20:48:41 +0000

I should have stated I'm using autoconf 2.57

"Dr. David Kirkby" wrote:
> 
> I keep getting this message if I use Sun's compiler (cc: Sun WorkShop
> 6 update 2 C 5.3 2001/05/15) on my Sun Ultra 80 with Solaris 9. I've
> set the environment variable CC to cc. If the CC variable is not set,
> then gcc is found and the bug does not appear. It might be a problem
> with my coding of configure.in, but since the configure script
> requests it is sent to address@hidden, I am doing this.
> 
> I'm not quite sure how a header file can be usable, but not present!!
> 
> If this email is not clear due to line wrapping, download version 1.20
> of the configure.in script from:
> 
> http://cvs.sourceforge.net/cgi-bin/viewcvs.cgi/atlc/sources/configure.in
> 
> checking pthreads.h usability... no
> checking pthreads.h presence... no
> checking for pthreads.h... no
> checking mpi.h usability... yes
> checking mpi.h presence... no
> configure: WARNING: mpi.h: accepted by the compiler, rejected by the
> preprocessor!
> configure: WARNING: mpi.h: proceeding with the preprocessor's result
> configure: WARNING:     ## ------------------------------------ ##
> configure: WARNING:     ## Report this to address@hidden ##
> configure: WARNING:     ## ------------------------------------ ##
> checking for mpi.h... no
> checking gsl/gsl_sf_ellint.h usability... yes
> checking gsl/gsl_sf_ellint.h presence... no
> configure: WARNING: gsl/gsl_sf_ellint.h: accepted by the compiler,
> rejected by the preprocessor!
> configure: WARNING: gsl/gsl_sf_ellint.h: proceeding with the
> preprocessor's result
> configure: WARNING:     ## ------------------------------------ ##
> configure: WARNING:     ## Report this to address@hidden ##
> configure: WARNING:     ## ------------------------------------ ##
> checking for gsl/gsl_sf_ellint.h... no
> checking gsl/gsl_sf_bessel.h usability... yes
> checking gsl/gsl_sf_bessel.h presence... no
> configure: WARNING: gsl/gsl_sf_bessel.h: accepted by the compiler,
> rejected by the preprocessor!
> configure: WARNING: gsl/gsl_sf_bessel.h: proceeding with the
> preprocessor's result
> configure: WARNING:     ## ------------------------------------ ##
> configure: WARNING:     ## Report this to address@hidden ##
> configure: WARNING:     ## ------------------------------------ ##
> checking for gsl/gsl_sf_bessel.h... no
> checking whether byte ordering is bigendian... yes
> checking for int... yes
> checking size of int... 4
> 
> sparrow /export/home/davek/atlc % cat configure.in
> dnl Process this file with autoconf to produce a configure script.
> 
> dnl Order of itmes is suggested in the autoconf documentation to be
> that
> dnl below, to be as below. Let's try to stick to it.
> 
> dnl AC_INIT(file)
> dnl checks for programs
> dnl checks for libraries
> dnl checks for header files
> dnl checks for typedefs
> dnl checks for structures
> dnl checks for compiler characteristics
> dnl checks for library functions
> dnl checks for system services
> dnl AC_OUTPUT([file...])
> 
> AC_INIT(atlc, 4.2.11-alpha, address@hidden)
> AM_INIT_AUTOMAKE(atlc, 4.2.11-alpha, address@hidden)
> 
> dnl Checks for programs.
> AC_PROG_CC
> AC_PROG_CXX
> AC_PROG_CXXCPP
> AC_PROG_RANLIB
> AC_ISC_POSIX
> 
> dnl AC_OUTPUT_COMMANDS([exit])
> 
> dnl Checks for libraries.
> AC_CHECK_LIB(m,main)
> AC_CHECK_LIB(c,memalign, AC_DEFINE(HAVE_MEMALIGN))
> 
> dnl check for optimised math library on SPARC
> dnl AC_CHECK_LIB(mopt,main)
> dnl check for the optimised
> dnl AC_CHECK_LIB(copt,main)
> 
> gsl_libgsl=1
> gsl_libcblas=1
> AC_CHECK_LIB(gslcblas,cblas_scasum,,gsl_libcblas=0)
> AC_CHECK_LIB(gsl,main,,gsl_libgsl=0)
> 
> dnl Checks for header files.
> 
> dnl AC_STDC_HEADERS
> AC_HAVE_HEADERS(stdio.h math.h errno.h stdlib.h string.h thread.h
> pthread.h pthreads.h mpi.h)
> gsl_ellint=1
> gsl_bessel=1
> AC_HAVE_HEADERS(gsl/gsl_sf_ellint.h,,gsl_ellint=0)
> AC_HAVE_HEADERS(gsl/gsl_sf_bessel.h,,gsl_bessel=0)
> 
> dnl checks for typedefs (none need to be tested)
> 
> dnl checks for structures (none need to be tested)
> 
> dnl checks for compiler characteristics
> AC_C_BIGENDIAN
> AC_CHECK_SIZEOF(int)
> AC_CHECK_SIZEOF(long)
> AC_CHECK_SIZEOF(size_t)
> 
> dnl checks for system services
> 
> dnl Check if the user wants a GUI. If so makes sure GTK and wxWindows
> are around.
> dnl The macro is in the file gtk.m4, so it must be read it here.
> sinclude(gtk.m4)
> AC_ARG_WITH(gui, AC_HELP_STRING([--with-gui=[no]],[Add a GUI (needs
> GTK >= 1.2.7 installed) but the GUI does not work!! ]))
> if test "x$with_gui" = "xyes"; then
>         AC_PATH_GTK(1.2.7, GTK127=1)
>         if test "$GTK127" != 1; then
>             AC_MSG_ERROR([
>                 Please check that gtk-config is in path, the directory
>                 where GTK+ libraries are installed (returned by
>                 'gtk-config --libs' command) is in LD_LIBRARY_PATH or
>                 equivalent variable and GTK+ is version 1.2.7 or
> above.
>             ])
>         fi
>         AC_CHECK_PROG(WXWINDOWS,wx-config, 1, 0)
>         if test "$WXWINDOWS" != 1; then
>             AC_MSG_ERROR([
>                wxWindows must be installed to use the GUI version of
> atlc
>         ])
>         fi
>         AC_SUBST(EXTRA_WXWINDOWS_LIBS,`wx-config --libs`)
>         AC_SUBST(EXTRA_WXWINDOWS_CFLAGS,`wx-config --cflags`)
>         AC_SUBST(GUI_DIRECTORY,gui)
> fi
> 
> dnl Multi-Processor Support
> AC_ARG_WITH(threads, AC_HELP_STRING([--with-threads=[no]],[support
> multiple processors (needs pthreads or similar installed)]))
> if test "x$with_threads" = "xyes"; then
> AC_DEFINE(ENABLE_MP)
> AC_DEFINE(_REENTRANT)
> AC_DEFINE(_THREAD_SAFE)
> AC_CHECK_LIB(thread,  pthread_create)
> AC_CHECK_LIB(pthread,  pthread_create)
> dnl thr_setconcurrency is only needed on Solaris as far as I know.
> AC_CHECK_LIB(thread, thr_setconcurrency,
> [AC_DEFINE(HAVE_THR_SETCONCURRENCY)])
> fi
> 
> dnl MPI and MPE Support
> AC_ARG_WITH(mpi,
> AC_HELP_STRING([--with-mpi=[mpi_install_prefix]],[support MPI-based
> parallel computation (needs MPICH installed)]))
> AC_ARG_WITH(mpe,
> AC_HELP_STRING([--with-mpe=[mpi_install_prefix]],[implies --with-mpi
> and additionally links in MPE logging support]))
> 
> if test "x$with_mpi" != "x" \
> || test "x$with_mpe" != "x"; then
> 
>   if test "x$with_mp" = "xyes"; then
>     AC_MSG_ERROR([it does not make sense to use --with-mp with
> --with-mpi or --with-mpe])
>   fi
> 
>   AC_DEFINE(ENABLE_MPI)
>   AC_CHECK_LIB(socket, getsockopt)
>   AC_CHECK_LIB(nsl, xdr_void)
>   AC_CHECK_LIB(rt, sched_yield)
>   if test "x$with_mpe" != "x"; then
>     if test "x$with_mpe" = "xyes"; then
>         path_to_mpe=/usr/local
>     else
>         path_to_mpe=$with_mpe
>     fi
>     CFLAGS="$CFLAGS -I$path_to_mpe/include -L$path_to_mpe/lib"
>     AC_CHECK_LIB(pmpich,  MPI_Init , , [AC_MSG_ERROR([Could not link
> test program with libpmpich.a])])
>     AC_CHECK_LIB(mpe,  MPE_Init_log , , [AC_MSG_ERROR([Could not link
> test program with libmpe.a])])
>     AC_CHECK_LIB(lmpe,  MPE_Init_log , ,[AC_MSG_ERROR([Could not link
> test program with liblmpe.a])] , -lmpe)
>   else
>     if test "x$with_mpi" = "xyes"; then
>         path_to_mpi=/usr/local
>     else
>         path_to_mpi=$with_mpi
>     fi
>     CFLAGS="$CFLAGS -I$path_to_mpi/include -L$path_to_mpi/lib"
>     AC_CHECK_LIB(mpich,  MPI_Init, , [AC_MSG_ERROR([Could not link
> test program with libmpich.a])])
>   fi
> fi
> 
> dnl Check for a functioning gsl installation.
> 
> if test $gsl_ellint = 1 && test $gsl_bessel = 1 && test $gsl_libcblas
> = 1 && test $gsl_libgsl = 1; then
>   AC_DEFINE(HAVE_GSL)
> elif test $gsl_ellint = 0 || test $gsl_bessel = 0 || test
> $gsl_libcblas = 0 || test $gsl_libgsl = 0; then
>   AC_MSG_WARN([To get full functionality you should install the GNU
>   scientific library (gsl) available at http://sources.redhat.com/gsl.
> Without
>   gsl it will be impossible to evaluate the theoretical impedance of a
> coupler
>   using \"design_coupler\" or of a stripline using
>   \"usage_create_bmp_for_symmetrical_stripline\". Some self tests will
> fail too.
>   However, much of the atlc package will work without gsl.])
> else
> dnl some libraries/headers exist, but not all of them.
>   AC_MSG_WARN([Some of the gsl headers and libraries were found, but
>   others were not. Since the installation appears to be broken, the
> use of
>   gsl will be disabled. Without gsl, it will be impossible to evaluate
> the
>   theoretical impedance of a coupler using \"design_coupler\" or of a
> stripline
>   using \"usage_create_bmp_for_symmetrical_stripline\". Some self
> tests will
>   fail too. However, much of the atlc package will work without gsl.])
> fi
> 
> AC_OUTPUT([\
> Makefile                                        \
> src/Makefile                                    \
> src/non_gui/Makefile                            \
> src/gui/Makefile                                \
> man/Makefile                                    \
> man/man1/Makefile                               \
> examples/Makefile                               \
> tools/Makefile                                  \
> tools/src/Makefile                              \
> tests/Makefile                                  \
> docs/html-docs/Makefile                         \
> docs/html-docs/jpgs/Makefile                    \
> docs/qex-december-1996/Makefile                 \
> docs/theory/Makefile                            \
> docs/Makefile                                   ])
> 
> --
> Dr. David Kirkby,
> Senior Research Fellow,
> Department of Medical Physics,
> University College London,
> 11-20 Capper St, London, WC1E 6JA.
> Tel: 020 7679 6408 Fax: 020 7679 6269
> Internal telephone: ext 46408
> e-mail address@hidden

-- 
Dr. David Kirkby,
Senior Research Fellow,
Department of Medical Physics,
University College London,
11-20 Capper St, London, WC1E 6JA.
Tel: 020 7679 6408 Fax: 020 7679 6269
Internal telephone: ext 46408
e-mail address@hidden




reply via email to

[Prev in Thread] Current Thread [Next in Thread]