One of the PLplot developers, Arjen Markus, has just reported to me that ctest -j4 for PLplot hangs on his MinGW-w64/MSYS2 platform while ctest works fine. This issue occurred for the mingw64/mingw-w64-x86_64-cmake 3.4.1-1 package version of ctest.exe, and the PLplot build was done with the "MSYS Makefiles" generator. In contrast to the MinGW-w64/MSYS2 platform case, ctest -j4 works fine for the Linux case.
Has anyone else here had similar difficulties OR success with the parallel ctest option on the MinGW-w64/MSYS2 platform? Previously, Arjen had discovered that the parallel build -j4 option for make.exe (from the msys/make 4.1-4 package) was not reliable on this platform. I have forgotten all the details for that issue, but we didn't pursue it further at that time because Cygwin also had the same parallel make trouble so we wrote it off to some make.exe bug for Cygwin that had been copied to MinGW-w64/MSYS2 platform (since the two platforms are so closely related). Of course, it always possible on both platforms that some make race condition caused by a badly implemented CMake-based build system for PLplot could be the source of these parallel make issues. However, make -j4 works fine for the Linux case so if there is such a make race issue it is a subtle one. To eliminate this possibility does there exist a definitive way to search for make race conditions for CMake-based builds? (I am thinking along the lines of looking for duplicate "Building ..." lines in the output from "make all", but there may be some better method that someone here would like to suggest.) Assuming make race conditions can be eliminated as the cause of the issue with the make parallel build option, is there any obvious connection between that issue and the parallel ctest issue on this platform? For example, is the ctest parallel build option implemented using generator facilities (e.g., the parallel build option for make.exe) or do both the ctest and make implementations of their respective parallel options use some common library that could be screwing up on the MinGW-w64/MSYS2 platform? 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 __________________________ -- Powered by www.kitware.com Please keep messages on-topic and check the CMake FAQ at: http://www.cmake.org/Wiki/CMake_FAQ Kitware offers various services to support the CMake community. For more information on each offering, please visit: CMake Support: http://cmake.org/cmake/help/support.html CMake Consulting: http://cmake.org/cmake/help/consulting.html CMake Training Courses: http://cmake.org/cmake/help/training.html Visit other Kitware open-source projects at http://www.kitware.com/opensource/opensource.html Follow this link to subscribe/unsubscribe: http://public.kitware.com/mailman/listinfo/cmake-developers