Alan,

I will do the run-time testing -- thank you for the detailed info!

I do have one more question: it seems that I do not have the static
libraries built under plplot (I only have the shared libraries -- can
fortran use these libraries?). Is there a flag I should set at
configuration time?

Thank you,

-- 
Irena


On Wed, Aug 27, 2014 at 12:15 PM, Alan W. Irwin <ir...@beluga.phys.uvic.ca>
wrote:

> To Arjen and Irena:
>
>
> On 2014-08-27 10:41-0400 Irena Johnson wrote:
>
>  Hi Arjen,
>>
>> Yes, this worked! I was now able to compile and install plplot for
>> pgf-13.6. Thank you very much for your help!
>>
>
> @Arjen: Please go ahead and push your workaround to the master
> branch of our git repository with a code comment that splitting up the
> module and subroutine this way is required to work around a pgf-13.6
> bug.  After all, that is a rather minor change, and others may want to
> use this version of the Portland Group compilers as well without
> having to upgrade to Davide's version 13.10.
>
> @Irena: that is excellent news indeed that Arjen's workaround solved
> the remaining build issue for you.  However, from Davide's results
> (where version 13.10 does not need this workaround) it sounds like the
> Portland Group version 13.6 compilers are a little iffy. Therefore I
> suggest to gain some confidence in results for those that you do some
> comprehensive run-time tests.  (You should run these tests
> every time you are using a new compiler or new compiler version.)
>
> The comprehensive test directions (I go into detail since they are
> suitable for anyone here who wants to test all is well with their
> PLplot build) are as follows:
>
> Add the cmake command-line option -DBUILD_TEST=ON which instructs
> cmake to configure many build-tree run-time tests.
>
> After cmake completes then run
>
> make -j4 VERBOSE=1 test_noninteractive >& test_noninteractive.out
>
> which builds and runs each of our 33 standard examples for all
> languages that are enabled plus a number of additional noninteractive
> tests.
>
> N.B. My high-quality but still 5-yr old PC takes ~8 minutes to
> run this test and produces some 4GB (!) of output plot files.
>
> You should check the resulting test_noninteractive.out file for any
> run-time errors.  (I just did that which confirms that our current git
> master version is working well.) And you can also sample some of those
> resulting PostScript files using gv, For example,
>
> gv --orientation=landscape examples/test_examples_output_dir/x08f95.psc
>
> views the PostScript results of our 8th standard example written
> in Fortran 95.
>
> An additional test that you should run is
>
> make -j4 VERBOSE=1 test_interactive >& test_interactive.out
>
> which tests all our interactive devices for our standard examples as
> well as many other interactive tests.  Unlike the test_noninteractive
> target, this target does not have to store plot results so does not
> chew up disk space.  But you do have to interact with a number of
> those examples (typically by hitting the enter key) to keep them
> moving along.  When you finish with this test, you should also check
> test_interactive.out for any run-time errors.
>
> Good luck with these essential comprehensive run-time tests, and
> please report back here if the resulting *.out files show any issues.
>
>
> 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
> __________________________
>
------------------------------------------------------------------------------
Slashdot TV.  
Video for Nerds.  Stuff that matters.
http://tv.slashdot.org/
_______________________________________________
Plplot-general mailing list
Plplot-general@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/plplot-general

Reply via email to