<div dir="ltr"><div>I applied the changes, and made a couple of my own, and now I think the integration works quite well. Thank you especially for the recursive implementation. The only problem I have is that I can't run individual tests (using "make perftest TESTS=somethin") from the top directory, as it also tries the m4/ subfolder, while the tests are only in src/. It works from src/ though, and I could make the python script just return when encountering a non-existent test. I made something similar in CMake, the target is not recursive since I don't think it's a prirority now, but it works. <br>
<br></div><div>And yes, I would like to move on to more interesting things. Even though time is running short, I would still like to do at least a simple server, with data upload and visualization from different computers. It can show the exact same things as the locally generated HTML, only from multiple computers/configurations for comparison. I also remember you mentioning some overview data, for a larger number of tests. What should such an overview tell us and how should it look like?<br>
</div></div><div class="gmail_extra"><br><br><div class="gmail_quote">2013/9/5 Christian Engwer <span dir="ltr"><<a href="mailto:christian.engwer@uni-muenster.de" target="_blank">christian.engwer@uni-muenster.de</a>></span><br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Hi Miha,<br>
<br>
you didn't reply to my suggested changes. Do they work as you<br>
intended? If yes, please apply them and we can continue with the<br>
interesting stuff ;-) buildsystem integration is an unlovely necessity.<br>
<br>
Ciao<br>
<span class="HOEnZb"><font color="#888888">Christian<br>
</font></span><div class="HOEnZb"><div class="h5"><br>
On Wed, Sep 04, 2013 at 01:52:07PM +0200, Miha Čančula wrote:<br>
> Hello all!<br>
><br>
> I'm realy sorry for being so unresponsive this week. I had some unforeseen<br>
> family matters, and I while I though I'd still be able to work on DUNE, I<br>
> couldn't. I'm back at work today, however, and it looks like I will stay<br>
> this way. I understand you're not happy about this, neither am I, but I<br>
> could hardly see it coming. I will do my best from now on, but I completely<br>
> understand if you feel that I haven't completed my project.<br>
><br>
> As I said, today I'm working again, and I'm just going through your<br>
> patches. I believe I added extra paths to $PATH on my computer for easier<br>
> development, and forgot to use the real paths in the generated code. The<br>
> last thing I did was completely change the arguments to perftest.py, so it<br>
> would be easier to call it from a Makefile using a single configuration<br>
> file. There are also some improvements to generated HTML, such as<br>
> navigation tabs. I will write a full report after I'm done with today's<br>
> work.<br>
><br>
> Thank you for being so patient with me :)<br>
><br>
> Regards, Miha<br>
><br>
><br>
> 2013/9/3 Christian Engwer <<a href="mailto:christian.engwer@uni-muenster.de">christian.engwer@uni-muenster.de</a>><br>
><br>
> > Hi,<br>
> ><br>
> > I have some patches here to integrate with the autotools (at least the<br>
> > way I think things are supposed to work) and to make OOS builds<br>
> > work. It seems a template file missing from the repositories.<br>
> ><br>
> > Have a look at the attached files and apply if it is what is intended :-)<br>
> ><br>
> > Cheers<br>
> > Christian<br>
> ><br>
> > On Tue, Sep 03, 2013 at 10:51:10PM +0200, Christian Engwer wrote:<br>
> > > Hi Miha,<br>
> > ><br>
> > > I must push a little bit. There hasn't been much progress this week<br>
> > > yet. End of GSOC is approaching and we still have some important steps<br>
> > > to do.<br>
> > ><br>
> > > What is the current state? YOur last blog entry is quite old and you<br>
> > > didn't post a recent status update on the list...<br>
> > > Please ask questions, otherwise we can not give help.<br>
> > ><br>
> > > I'll try to look through your latest changes :-)<br>
> > ><br>
> > > Ciao<br>
> > > Christian<br>
> > ><br>
> > > On Thu, Aug 29, 2013 at 10:54:22AM +0200, Miha Čančula wrote:<br>
> > > > Yes, I probably should update it, now that the behavior is more or less<br>
> > > > fixed.<br>
> > > ><br>
> > > > Unfortunately, I've been a little stuck on the buildsystem integration.<br>
> > > > This is necessary for getting the compiler and compile flags, and it's<br>
> > also<br>
> > > > more user-friendly. I imagined it like this:<br>
> > > > 1.) At configure-time, a configuration file (currently named<br>
> > "<name>.ini",<br>
> > > > can be changed if needed) is generated for each program you want to<br>
> > measure<br>
> > > > 2.) Additionally, a make target ("perftest_<name>") is created, so you<br>
> > can<br>
> > > > just call "make perftest_myprogram"<br>
> > > > 3.) There is another target, called just "perftest", which runs all the<br>
> > > > perftest_* targets it can find<br>
> > > ><br>
> > > > Such a system seems quite reasonable for me, because most of the time<br>
> > the<br>
> > > > user will just call "make perftest" and be done with it. I implemented<br>
> > the<br>
> > > > exact same system in CMake. However, when I try to do the same thing<br>
> > using<br>
> > > > autotools, I can't seem to figure out how to do the first step. I have<br>
> > a<br>
> > > > template (<a href="http://perftest.ini.in" target="_blank">perftest.ini.in</a>), which I would like to replicate multiple<br>
> > times<br>
> > > > using ac_config_file(), but each time with different variables. Calling<br>
> > > > ac_config_file() and ac_output() multiple times in various combination<br>
> > > > doesn't work, it always uses the last values. You probably have more<br>
> > > > experience with autotools, is there an easier way to do it?<br>
> > > ><br>
> > > > If there isn't, I can of course write another script (perhaps in<br>
> > Python)<br>
> > > > that generates the configuration file, but that would introduce even<br>
> > more<br>
> > > > complexity.<br>
> > > ><br>
> > > > I had creating a single overview HTML file on the agenda for this<br>
> > week, and<br>
> > > > I think this would be easier, but I wanted to finish the buildsystem<br>
> > > > integration first. Due to the problems I think I'll do the overview<br>
> > first,<br>
> > > > update the documentation, and then try the buildsystem again.<br>
> > > ><br>
> > > ><br>
> > > > 2013/8/27 Christoph Grüninger <<br>
> > <a href="mailto:christoph.grueninger@iws.uni-stuttgart.de">christoph.grueninger@iws.uni-stuttgart.de</a>><br>
> > > ><br>
> > > > > Hi Miha,<br>
> > > > > could you be so kind and update the README, so that the barrier to<br>
> > test<br>
> > > > > your module is as low as possible? You should add that numPy is an<br>
> > optional<br>
> > > > > dependency, too.<br>
> > > > ><br>
> > > > > What's on your agenda for this week?<br>
> > > > ><br>
> > > > > Bye<br>
> > > > > Christoph<br>
> > > > ><br>
> > > > ><br>
> > > > > Miha Čančula <<a href="mailto:miha@noughmad.eu">miha@noughmad.eu</a>> schrieb:<br>
> > > > > > Hello, Christoph!<br>
> > > > > ><br>
> > > > > > I changed the behavior of the script, the argument should not be<br>
> > the<br>
> > > > > actual<br>
> > > > > > executable to run, but rather a configuration file that describes<br>
> > what to<br>
> > > > > > measure and some extra data. Try<br>
> > > > > ><br>
> > > > > > python3 perftest.py ../../src/dune_perftest_matrix.ini<br>
> > > > > ><br>
> > > > > > The configuration files are generated as part the build process,<br>
> > if you<br>
> > > > > use<br>
> > > > > > autoconf they're located in src, if you use cmake they're in<br>
> > > > > > build-cmake/src. You can alternatively call it by running 'make<br>
> > > > > > src/perftest_dune_perftest_matrix', or just 'make perftest'. The<br>
> > > > > generation<br>
> > > > > > of files is still not completely correct (you get some wrong names<br>
> > with<br>
> > > > > > multiple output files), but it should work.<br>
> > > > ><br>
> > ><br>
> > > > _______________________________________________<br>
> > > > Dune-devel mailing list<br>
> > > > <a href="mailto:Dune-devel@dune-project.org">Dune-devel@dune-project.org</a><br>
> > > > <a href="http://lists.dune-project.org/mailman/listinfo/dune-devel" target="_blank">http://lists.dune-project.org/mailman/listinfo/dune-devel</a><br>
> > ><br>
> > ><br>
> > > --<br>
> > > Prof. Dr. Christian Engwer<br>
> > > Institut für Numerische und Angewandte Mathematik<br>
> > > Fachbereich Mathematik und Informatik der Universität Münster<br>
> > > Einsteinstrasse 62<br>
> > > 48149 Münster<br>
> > ><br>
> > > E-Mail <a href="mailto:christian.engwer@uni-muenster.de">christian.engwer@uni-muenster.de</a><br>
> > > Telefon <a href="tel:%2B49%20251%2083-35067" value="+492518335067">+49 251 83-35067</a><br>
> > > FAX <a href="tel:%2B49%20251%2083-32729" value="+492518332729">+49 251 83-32729</a><br>
> > ><br>
> > > _______________________________________________<br>
> > > Dune-devel mailing list<br>
> > > <a href="mailto:Dune-devel@dune-project.org">Dune-devel@dune-project.org</a><br>
> > > <a href="http://lists.dune-project.org/mailman/listinfo/dune-devel" target="_blank">http://lists.dune-project.org/mailman/listinfo/dune-devel</a><br>
> ><br>
> > --<br>
> > Prof. Dr. Christian Engwer<br>
> > Institut für Numerische und Angewandte Mathematik<br>
> > Fachbereich Mathematik und Informatik der Universität Münster<br>
> > Einsteinstrasse 62<br>
> > 48149 Münster<br>
> ><br>
> > E-Mail <a href="mailto:christian.engwer@uni-muenster.de">christian.engwer@uni-muenster.de</a><br>
> > Telefon <a href="tel:%2B49%20251%2083-35067" value="+492518335067">+49 251 83-35067</a><br>
> > FAX <a href="tel:%2B49%20251%2083-32729" value="+492518332729">+49 251 83-32729</a><br>
> ><br>
<br>
--<br>
Prof. Dr. Christian Engwer<br>
Institut für Numerische und Angewandte Mathematik<br>
Fachbereich Mathematik und Informatik der Universität Münster<br>
Einsteinstrasse 62<br>
48149 Münster<br>
<br>
E-Mail <a href="mailto:christian.engwer@uni-muenster.de">christian.engwer@uni-muenster.de</a><br>
Telefon <a href="tel:%2B49%20251%2083-35067" value="+492518335067">+49 251 83-35067</a><br>
FAX <a href="tel:%2B49%20251%2083-32729" value="+492518332729">+49 251 83-32729</a><br>
</div></div></blockquote></div><br></div>