[cmake-developers] Incomplete gfortran library link command sometimes mysteriously occurs with MinGW/MSYS on Wine-1.5.15 platform--SOLVED

[Alan W. Irwin]

On 2012-10-26 16:24-0700 Alan W. Irwin wrote:

> On 2012-10-26 02:15-0700 Alan W. Irwin wrote:
>
>> On 2012-10-25 23:26-0600 Clinton Stimpson wrote:
>> 
>>> Alan, you are in luck.  They fixed it two days ago!!  At first, I was 
>>> confused because I was not able to reproduce it, then trying it against an 
>>> older version of Wine showed the problem.
>>> 
>>> commit 4a566ed3d6acd8a6e18eeaeb41d55d0f793029de
>>> Author: Piotr Caban <piotr at codeweavers.com>
>>> Date:   Tue Oct 23 18:11:06 2012 +0200
>>>
>>>    msvcp90: Convert characters to unsigned char before passing them to 
>>> functions operating on integers.
>>> 
>> 
>> That's awesome.
>
> Furthermore, I just discovered that patch got into the just-released
> Wine-1.5.16 so I will give that version a try for my comprehensive checks
> rather than attempting to carry on with a patched version of 1.5.15.

I delayed doing a comprehensive check because of other issues such as
the backwards incompatibility of language support with cmake-2.8.10
that I had to deal with + a recent Wine fix where I helped find the
solution with extensive testing. That fix consisted of a complete
solution to the long-standing Wine issue with scanf precision
(http://bugs.winehq.org/show_bug.cgi?id=28422) that made it into
wine-git yesterday thanks to Piotr Caban's efforts as well as my
testing.  I also discovered yesterday that due to some other recent
fix (not available for wine-get-2012-11-02) one of my longer tests
completed 5 (!) times quicker than previously.  That test on Wine
still takes a factor of two longer to complete than the equivalent
Linux test, but I was nevertheless quite happy to see that enormous
Wine speed improvement.

Today I was finally able to finish those tests which consisted of
complete builds and tests for shapelib (a prerequisite of PLplot),
PLplot (a prerequisite of the many plotting tests for te_gen), ephcom,
and te_gen.  The platform consisted of CMake-2.8.10 (the downloaded
version rather than a bootstrapped version), MinGW-4.7.0 (installed
with the automatic installer if you choose the update option), MSYS,
and wine-git-2012-11-05.  Except for one issue with our private Lua
find module for PLplot which I haven't looked into yet, all the
extremely extensive test results on that platform were perfect.  So I
would definitely recommend trying that platform (or Wine-1.5.17 for
the Wine component of that platform when it gets released this
weekend) for anybody here interested in doing builds and tests of their
software on a free (in both senses) Windows platform.

IOW, I think using the CMake/MinGW/MSYS/Wine platform as a Windows
testing platform has moved from the crazy early adopter stage where
only one person (i.e., me) was trying this to the early adopter stage
where a larger group (who are still don't mind living on the edge)
starts trying it.

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
__________________________