Federico Perini

Other Fortran libraries such as [json-fortran](https://github.com/jacobwilliams/json-fortran) already implement these tools and could serve as an inspiration for the CI setup

Also: it would be worthwhile to check whether the test programs are run with OpenMP turned on, as it it known that stdlib is still not 100% thread-safe

I think it's a nice idea @coriollis-nico, it could be made part of the `[fortran]` features table in `fpm.toml.`.

I think it's a great idea @awvwgk - I will ping @wassup05 in case that maybe a suitable contribution for his GSoC project. Would this implementation be based on a...

It seems like all comments have been addressed. Let's wait one or two days for further comments, and then merge.

Because the change only affects test routines, and the ongoing onstability of said tests, I am in favor of the PR proposal, although it's not great coding practice. @jvdp1 what...

One way to identify the OS and save it as a `parameter` would be to use the `compiler_version` and `compiler_options` intrinsic functions. It will certainly need quite some testing and...

[v8](https://github.com/fortran-lang/setup-fpm/releases/tag/v8) is compatible with arm64 macs, and is in the CI/CD pipeline already. Of course for past releases/those where a binary is not available, it must be built from sources....

Thanks for the hint @rouson - I have tested your exact issue but cannot reproduce it on macOS: everything builds OK. However, if I set this instead (note, lowercase dependency...

Thank you for posting this @davidpfister - Looking at the code, I see the `fpm_build_settings` are [properly validated](https://github.com/fortran-lang/fpm/blob/501cc61ef1b37a4a875dcd266e0ba8cedebf2fcf/src/fpm_command_line.f90#L1619-L1621) on CLI initialization, which is the expected behavior. However, you may be...