[C++-sig] Boost.Python, AIX, and gcc

J. Michael Owen mikeowen at llnl.gov
Wed Nov 5 20:29:26 CET 2003


I just updated to the lastest boost CVS, and there's one slight typo that crept
into the AIX gcc build.  It's in file tools/build/v1/python.jam, and here's the
diff with the appropriate fix.

Index: tools/build/v1/python.jam
===================================================================
RCS file: /cvsroot/boost/boost/tools/build/v1/python.jam,v
retrieving revision 1.65
diff -u -r1.65 python.jam
--- tools/build/v1/python.jam	5 Nov 2003 04:49:10 -0000	1.65
+++ tools/build/v1/python.jam	5 Nov 2003 19:25:32 -0000
@@ -124,7 +124,7 @@
     else if $(OS) = AIX
     {
         PYTHON_PROPERTIES 
-          += <*><*><linkflags>"-Wl,bI:$(PYTHON_LIB_PATH)/python.exp"
+          += <*><*><linkflags>"-Wl,-bI:$(PYTHON_LIB_PATH)/python.exp"
             <*><*><find-library>pthreads ;
     }
 }

On Monday 03 November 2003 03:42 pm, Rene Rivera wrote:
> [2003-11-03] J. Michael Owen wrote:
> >I've been trying to use the latest Boost.Python with gcc on AIX, which is
>
> always
>
> >fun with AIX's strange dynamic library system.  I now have it working
> > after modifying the some of the jam build files, and one source file. 
> > The files
>
> I've
>
> >altered are:
> >
> >tools/build/v1/boost-base.jam
> >tools/build/v1/gcc-tools.jam
> >tools/build/v1/python.jam
> >boost/type_traits/type_with_alignment.hpp
>
> OK, but a diff of the changes would be way more usefull ;-) As we could
> look at them without having to figure what changes you made are.
>
>
> -- grafik - Don't Assume Anything
> -- rrivera (at) acm.org - grafik (at) redshift-software.com
> -- 102708583 (at) icq

-- 
 "Hey...where are the sunflower seeds?" |       J. Michael Owen         
        o_o /                           |       Phone:  925-423-7160     
        (")                             |       Email:  mikeowen at llnl.gov
       \/'\/                            |
____(__(,_,)_______________________________________________________________




More information about the Cplusplus-sig mailing list