On 2011-11-10 08:49+0100 Arjen Markus wrote:

> i Alan,
>
> On 2011-11-09 19:19, Alan W. Irwin wrote:
>
>>
>> Is this the first time you have tested the MinGW version of gfortran
>> on Windows for our f95 examples using the "MinGw Makefiles" generator?
>> As I recall, I could not get those examples to work on MinGW/Wine for
>> the "MSYS Makefiles" generator, but I believe it was a separate issue
>> (complaints about the form of the module file).  I wrote that off as a
>> Wine platform problem and probably didn't get deep enough into the
>> testing to run into the problem you describe.
>>
>
> No, the MinGW+gfortran combination (not MSYS) is the one I use most.
> It is only very recent that I noticed these issues.

Have you ever tested the combination of CMake-2.8.x and MinGW before?
I just realized that the -Wl,--allow-multiple-definition option has
been automatically set in our
cmake/modules/language_support/cmake-2.6/Platform/*.cmake files that
override CMake-2.6 language support in the Windows case. If you
previously were just using CMake-2.6.x to test this case, then you
were always using the -Wl,--allow-multiple-definition option which
would explain why you never saw this issue before.

However, if it turns out you have successfully tested CMake-2.8.x and
MinGW before, then it is important to figure out exactly what has
changed between the success you had before and now.  For example, if
you can show an older version of CMake-2.8.x works but not
CMake-2.8.6, that is important information that should be passed on to
the CMake developers.

You have recently copied 5 Fortran language support files from
cmake/modules/language_support/cmake-2.6 to
cmake/modules/language_support/cmake-2.8 to propagate the
-Wl,--allow-multiple-definition to the CMake-2.8.x case. However, the
-Wl,--allow-multiple-definition option is just a brute-force
workaround for a more fundamental issue so I hope one of the
suggestions I made previously (e.g., installing the latest MinGW
version of gfortran) will provide a more fundamental solution to that
issue.  That would allow you to permanently remove those 5 CMake-2.6.x
Fortran language support files in
cmake/modules/language_support/cmake-2.8 and therefore give Windows
users the full benefits of the improved Fortran language support in
CMake-2.8.x.  Of course, in any case you should remove those files
locally for all your further experiments.

Good luck with your further CMake-2.8.6 experiments
to find a fundamental way to avoid having to use the
-Wl,--allow-multiple-definition workaround.

Alan

__________________________
Alan W. Irwin

Astronomical research affiliation with Department of Physics and Astronomy,
University of Victoria (astrowww.phys.uvic.ca).

Programming affiliations with the FreeEOS equation-of-state
implementation for stellar interiors (freeeos.sf.net); the Time
Ephemerides project (timeephem.sf.net); PLplot scientific plotting
software package (plplot.sf.net); the libLASi project
(unifont.org/lasi); the Loads of Linux Links project (loll.sf.net);
and the Linux Brochure Project (lbproject.sf.net).
__________________________

Linux-powered Science
__________________________

------------------------------------------------------------------------------
RSA(R) Conference 2012
Save $700 by Nov 18
Register now
http://p.sf.net/sfu/rsa-sfdev2dev1
_______________________________________________
Plplot-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/plplot-devel

Reply via email to