gcl-devel
[Top][All Lists]
Advanced

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

[Gcl-devel] Re: GCL results on Solaris


From: Camm Maguire
Subject: [Gcl-devel] Re: GCL results on Solaris
Date: 20 Dec 2004 11:22:29 -0500
User-agent: Gnus/5.09 (Gnus v5.9.0) Emacs/21.2

Greetings!  And thanks for your feedback!

The problem appears to be, from the make log you sent, that the local
bfd library is not being built at all.  Either this, or it has been
built in a previous run and not cleaned before this one.  The latter
is actually the more probable, as configure did setup the BUILD_BFD
make direction correctly.  What you should see immediately after
configure ends and make begins is:

add-defs1 386-linux
using 386-linux.defs
cd binutils/bfd && make
make[1]: Entering directory `/fix/t1/camm/debian/gcl/gcl-2.6.5/binutils/bfd'

So lets assume you already have the h/bfd.h and binuntils/bfd/libbfd.a
(et.al.).

Even if you don't have boolean defined in your gcc (quite doubtful),
you should get it in h/bfd.h generated by the local bfd build:

#if defined (__GNUG__) && (__GNUC__ > 2 || (__GNUC__ == 2 && __GNUC_MINOR__ >= 
6))
#define TRUE_FALSE_ALREADY_DEFINED
#else
#if defined (__bool_true_false_are_defined)
/* We have <stdbool.h>.  */
#define TRUE_FALSE_ALREADY_DEFINED
#endif
#endif
#ifdef MPW
/* Pre-emptive strike - get the file with the enum.  */
#include <Types.h>
#define TRUE_FALSE_ALREADY_DEFINED
#endif /* MPW */
#ifndef TRUE_FALSE_ALREADY_DEFINED
typedef enum bfd_boolean {false, true} boolean;
#define BFD_TRUE_FALSE
#else
/* Use enum names that will appear nowhere else.  */
typedef enum bfd_boolean {bfd_fffalse, bfd_tttrue} boolean;
#endif

Once you have this file, you might try the simple test program:

#include "h/bfd.h"
int main() {
boolean x;
return 0;
}

If this compiles, then we're left with the following -- I notice
you've placed /opt/csw/include in the -I flags as opposed to the
C_INCLUDE_PATH environment variable.  The latter is more flexible, and
also differs from what you've done insofar as your method preempts the
standard -I gcc command line flags (-I../h -I../gcl-tk) as well as the
system include path.  If you have a bfd.h in /opt/csw/include, it will
preempt h/bfd.h, and may lead to this problem, but will certainly lead
to other problems later if not.  bfd.h and libbfd.a/libiberty.a must
be synchronous.

So my first suggestion is to make sure the gcl directory is cleaned to
start (make clean in this directory), remove the -I/opt.... flag, do
'export C_INCLUDE_PATH=/opt/csw/include' (or setenv C_INCLUDE_PATH
/opt... if using csh) as well as the analogous settings for
LIBRARY_PATH, LD_LIBRARY_PATH, and possibly PATH as posted previously,
retry, and send me the results again exactly as you did so helpfully
this time.  

Please feel free to write back if any of this needs clarification --
I'm writing rather quickly.  Also, will be leaving town for about 1
week on Tue. evening.

Take care,

"Kostas Oikonomou" <address@hidden> writes:

> Hi Camm,
> 
> Thanks for all your comments and hints.
> 
> Unfortunately, I'm still unable to get the GCL compilation to work.
> 
> Here's what I did:
> 
> 1. I applied all the patches you sent me.
> 2. I used --disable-statsysbfd --enable-locbfd.
> (I had tried this before, and I reported the results in the first email I 
> sent you:
> the make fails when compiling bfd_combined_table_update.)
> 3. This time, having read your comments on HAVE_BFD_BOOLEAN, I edited
> the configure script and added #undef HAVE_BFD_BOOLEAN.  If I understood
> what you said, this would be the correct thing to do.  I also made one other 
> small change
> to configure, but I don't think it's important for the present problem.
> 
> Despite these changes, the build failed in exactly the same way.
> 
> Here is a .tgz file containing the output of the make until the failure, my 
> Makefile.pamphlet,
> my lsp/Makefile.pamphlet, and my lsp/gcl-2.6.5/configure.
> 
> I hope you can figure out what's going wrong.  Please tell me if you need 
> more information.
> Again, thanks very much for your assistance.
> 
>                                       Kostas
> 

-- 
Camm Maguire                                            address@hidden
==========================================================================
"The earth is but one country, and mankind its citizens."  --  Baha'u'llah




reply via email to

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