These checked that e.g. the cpp and fc ids are identical, which isn't
strictly what we want. Particularly, msvc doesn't even have a fortran
compiler, and what we really care about is whether we mix both gcc and
something else.
This should be done in all cases of language_stdlib_only_link_flags, but
I don't have access to all of the compilers to test it.
This is required in cases where object files created by gfortran are
linked using another compiler with a differen default search path, such
as gfortran and clang together.
Since we changed to using a json file to avoid over long command lines
we created a situation where the generated files may not be ready when
the depscan happens. To avoid that, we need to add all of the generated
sources as order deps.
Fixes: #9258
This adds a test which makes use of an install:yes static library that
depends on another static library. This triggers a promotion to
link_whole_target inside meson which takes different code paths in
certain places.
Also makes use of .F90 source (capital F) to test for case
(in)sensitivity.
D lang compilers have an option -release (or similar) which turns off
asserts, contracts, and other runtime type checking. This patch wires
that up to the b_ndebug flag.
Fixes#7082
Fortran: check for undeclared variables by forcing implicit none everywhere
C/C++: check for unused parameters and return types
removed unused variables from test cases
ci: do missing return and unused arg check with Github Actions
* intel-cl tests: more rigorous detection of intent to use Intel Windows compilers
* fortran coarray test: make skipping more robust in that underlying MPI stack is .run()
This is useful for any Fortran coarray work, and especially for intel-cl where multiple Intel compiler
versions are often installed, and the wrong underlying MPI library may be dynamically linked,
and so a runtime check is needed to exercise the MPI stack underlying Fortran coarray.
This is done by
fc.run('sync all; end', dependencies: coarray)
* pep8
Apparently we have no tests for this because this is broken pretty
badly. This extends the basic test to actually check for the correct
free-form argument and thus test this.
Volker Weißmann
Eli Schwartz
Dylan Baker
Rohit Goswami
Mat Cross
Daniel Mensinger
Michael Hirsch
Jussi Pakkanen
Jonas Lundholm Bertelsen
Nirbheek Chauhan
Michael Hirsch, Ph.D