problems with gfortran module creation and "make -j"

[cyrush]

It seems the compiler is undermining threaded builds:

Fatal Error: Can't rename module file '../../../fortran/f_conduit_node.mod0' to '../../../fortran/f_conduit_node.mod': No such file or directory
make[2]: *** [tests/conduit/fortran/CMakeFiles/f_conduit_node_obj.dir/f_conduit_node_obj.f.o] Error 1
make[1]: *** [tests/conduit/fortran/CMakeFiles/f_conduit_node_obj.dir/all] Error 2
make[1]: *** Waiting for unfinished jobs....
[ 76%] /private/var/folders/wj/jjhjt_0d3kg1vmk56d5bq_38001kr8/T/spack-stage/spack-stage-J_e5zR/conduit/src/tests/conduit/fortran/f_conduit_node.f:411.6:

use f_conduit_node
1
Fatal Error: Can't open module file 'f_conduit_node.mod' for reading at (1): No such file or directory
make[2]: *** [tests/conduit/fortran/CMakeFiles/f_conduit_node.dir/f_conduit_node.f.o] Error 1

http://stackoverflow.com/questions/29728867/cmake-parallel-build-for-fortran-90-with-modules

[cyrush]

I think the problem is cropping up in unit testing. The solution may simply be to create a lib for fruit AND the fortran gtest driver.

[cyrush]

This could also bug specific to gfortran 4.9 and 4.9.1
http://stackoverflow.com/questions/27281158/fortran-90-gfortran-fatal-error-cant-rename-module-file-exists
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=62215

[cyrush]

closing, I don't see this any more on OSX 10.11 using gfortran 6.1. Assuming this was resolved by the compiler update.

[cyrush]

seeing with gfortran 5 (or even 7) so this isn't just the 4.9 bug.

Seems like a chicken vs egg issue. Object libs might help, or setting more fortran flags:

https://stackoverflow.com/questions/43917443/fortran-module-files-not-found-by-cmake

[cyrush]

https://gitlab.kitware.com/cmake/cmake/issues/17525

[cyrush]

Try not using a single Fortran_MODULE_DIRECTORY dir, by tweaking
SetupFortran.cmake:

    set(CMAKE_Fortran_MODULE_DIRECTORY ${PROJECT_BINARY_DIR}/fortran)
      
      # make sure the fortran compiler can see the module files it
      # generates
      include_directories(${CMAKE_Fortran_MODULE_DIRECTORY})

[cyrush]

finally got to the bottom of this.

It was a very simple error: I had a copy and paste typo in one of the test modules leading to a name collision.

The t_conduit_node_obj test, was defining a module t_conduit_node instead of t_conduit_node_obj. Since the t_conduit_node test also defines the t_conduit_node module -- they would stomp on each other when compiled in parallel.