Hello Rob, I don't have access to MSVC but just tested building cxx11_tensor_broadcasting on Ubuntu and macOS and got the following results:
macOS (Apple LLVM version 10.0.1 (clang-1001.0.46.4)): ./buildtests.sh cxx11_tensor_broadcasting 3.92s user 0.22s system 96% cpu 4.281 total Ubuntu (gcc version 5.4.0): ./buildtests.sh cxx11_tensor_broadcasting 5,57s user 0,41s system 99% cpu 6,037 total Running the whole Eigen test suite usually takes some time but I cannot reproduce your extreme build time on my systems. Cheers, David > On 7. Jun 2019, at 20:39, Rob Conde <[email protected]> wrote: > > So i'm building the tests on windows using vs2017 c++17 and finding that > building the tests is taking a LOOOOOONNNNGGG time. > cxx11_tensor_broadcasting.cpp took 2 hours to compile alone. Is this expected? > > Rob > > BTW I did notice at first that cmake will use the 64-bit cross tools (i.e. a > 32-bit cl which builds 64-bit binaries) by default, but i've already fixed > that problem. > From: Rob Conde <[email protected]> > Sent: Thursday, June 6, 2019 3:11 PM > To: [email protected] > Subject: Re: [eigen] 3.3 branch state > > Hey Christoph, > > According to the dashboard: > http://manao.inria.fr/CDash/index.php?project=Eigen > <http://manao.inria.fr/CDash/index.php?project=Eigen> > > 2-3 tests are failing currently. Is there any context to know if this is > acceptable? > > Thanks, > Rob Conde > From: Christoph Hertzberg <[email protected]> > Sent: Friday, May 31, 2019 10:10 AM > To: [email protected] > Subject: Re: [eigen] 3.3 branch state > > > > On 31/05/2019 15.48, Rob Conde wrote: > > I apologize, I got myself a bit confused (too much multi-tasking). Let's > > start over. > > > > According to Gael on the "About dropping C++03 compatibility" thread: > > > > this is off topic but with c++17 and gcc 7+ or clang 7+ and the head of > > Eigen, it was already possible to get rid of > > EIGEN_MAKE_ALIGNED_OPERATOR_NEW in user code. > > > > As of today, this is also conditionally removed from Eigen's classes and, > > more importantly, documented: > > http://eigen.tuxfamily.org/dox-devel/group__TopicUnalignedArrayAssert.html > > <http://eigen.tuxfamily.org/dox-devel/group__TopicUnalignedArrayAssert.html> > > > > > > That's the change I want...so I care not about 3.3, but master (...on > > github, default on bitbucket). So then the question is about the the state > > of master. If the answer is "run the tests yourself", that's perfectly > > acceptable. But if, for example, you run them before merging to > > default/master (and therefore default/master can be considered "blessed") > > then maybe I don't need to bother. > > For default/master I (and occasionally some other users) also run > Nightly builds: > http://manao.inria.fr/CDash/index.php?project=Eigen > <http://manao.inria.fr/CDash/index.php?project=Eigen> (also for C++03 and > C++11 only. Eventually, I'm planning to make running with newer > standards more easy. See http://eigen.tuxfamily.org/bz/show_bug.cgi?id=1655 > <http://eigen.tuxfamily.org/bz/show_bug.cgi?id=1655>) > > Gaël also runs builds which are triggered by any incoming commit (I > think also with C++14 or C++17, but I'm not sure). > "Usually" any changes are tested before they are pushed, but accidents > can happen, of course. And naturally, there can also be errors which are > not (yet) covered by any unit test. > > On the other hand, nothing (implicitly) happens due to failed tests (in > fact some failing tests are known issues ...), unless someone looks at > the dashboard and takes care of them (I usually do, but there is no > guarantee for that). > > If you want to be sure, pull some version, run the test-suite yourself > and only update occasionally (e.g., if you need a newly added feature or > bug-fix). And please report any issues you get! > > > Cheers, > Christoph > > > > > > Thanks! > > Rob > > ________________________________ > > From: Christoph Hertzberg <[email protected]> > > Sent: Friday, May 31, 2019 6:00 AM > > To: [email protected] > > Subject: Re: [eigen] 3.3 branch state > > > > I do make nightly test-runs for various gcc/clang versions for Eigen > > 3.3, but only using C++03 and C++11: > > http://manao.inria.fr/CDash/index.php?project=Eigen+3.3 > > <http://manao.inria.fr/CDash/index.php?project=Eigen+3.3> > > > > More extensive tests are usually done before releases. But the 3.3 > > branch generally only gets bug-fixes, so usually no new bugs should get > > introduced. > > Feel free to run the test-suite for your platform/compiler before using > > it (running the test-suite is relatively easy): > > > > I do not know what exactly you are referring to by "re: > > Fixed-size-vectorizable-types" (I can't find any mail, commit, > > pull-request or bugzilla entry with that title) > > > > > > Cheers, > > Christoph > > > > > > On 30/05/2019 18.27, Rob Conde wrote: > >> I want to grab the 3.3 branch head to get the C++17 updates re: > >> Fixed-size-vectorizable-types. Can I assume that all the tests are > >> passing, or is that something you only check when you do a release? > >> > >> Thanks, > >> Rob Conde > >> > > > > -- > > Dr.-Ing. Christoph Hertzberg > > > > Besuchsadresse der Nebengeschäftsstelle: > > DFKI GmbH > > Robotics Innovation Center > > Robert-Hooke-Straße 5 > > 28359 Bremen, Germany > > > > Postadresse der Hauptgeschäftsstelle Standort Bremen: > > DFKI GmbH > > Robotics Innovation Center > > Robert-Hooke-Straße 1 > > 28359 Bremen, Germany > > > > Tel.: +49 421 178 45-4021 > > Zentrale: +49 421 178 45-0 > > E-Mail: [email protected] > > > > Weitere Informationen: http://www.dfki.de/robotik > > <http://www.dfki.de/robotik> > > ------------------------------------------------------------- > > Deutsches Forschungszentrum für Künstliche Intelligenz GmbH > > Trippstadter Strasse 122, D-67663 Kaiserslautern, Germany > > > > Geschäftsführung: > > Prof. Dr. Jana Koehler (Vorsitzende) > > Dr. Walter Olthoff > > > > Vorsitzender des Aufsichtsrats: > > Prof. Dr. h.c. Hans A. Aukes > > Amtsgericht Kaiserslautern, HRB 2313 > > ------------------------------------------------------------- > > > > > > > > > > -- > Dr.-Ing. Christoph Hertzberg > > Besuchsadresse der Nebengeschäftsstelle: > DFKI GmbH > Robotics Innovation Center > Robert-Hooke-Straße 5 > 28359 Bremen, Germany > > Postadresse der Hauptgeschäftsstelle Standort Bremen: > DFKI GmbH > Robotics Innovation Center > Robert-Hooke-Straße 1 > 28359 Bremen, Germany > > Tel.: +49 421 178 45-4021 > Zentrale: +49 421 178 45-0 > E-Mail: [email protected] > > Weitere Informationen: http://www.dfki.de/robotik > <http://www.dfki.de/robotik> > ------------------------------------------------------------- > Deutsches Forschungszentrum für Künstliche Intelligenz GmbH > Trippstadter Strasse 122, D-67663 Kaiserslautern, Germany > > Geschäftsführung: > Prof. Dr. Jana Koehler (Vorsitzende) > Dr. Walter Olthoff > > Vorsitzender des Aufsichtsrats: > Prof. Dr. h.c. Hans A. Aukes > Amtsgericht Kaiserslautern, HRB 2313 > -------------------------------------------------------------
