Sylwester Arabas

Especially as installation of Assimulo from pip is broken at least since Dec 2019 (https://github.com/modelon-community/Assimulo/issues/7)

Thank you @bdcbqa314159! We now also have CI script for Github Actions, and it actually helped here - there seem to be some compilation errors, could you please have a...

> Do you have a framework to test code and see things in your personal computer or you only rely here? The two test suite calls: https://github.com/blitzpp/blitz/blob/f24a250a43dff88c31ad92916da828b7ea9a98b7/.github/workflows/main.yml#L34-L35 is what we...

voting up! Something like CUDA's atomicAdd, atomicMin, etc would be helpful. For the OMP hackend, the ``#pragma omp atomic`` should be of some help?

would this be a good place to issue such warning: https://github.com/numba/numba/blob/e15cd6870042125ff025e459f5f7558449b22cf8/numba/core/typed_passes.py#L338 (just after the logic for issuing warning for `parallel=True` with no parallelizable loops)

FWIW, here's another simple ctypes-based solution that works within njitted code, tested on Linux, Windows and macOS (https://github.com/open-atmos/PyMPDATA/blob/main/PyMPDATA/impl/clock.py): ```Python """ CPU-time returning clock() function which works from within njit-ted code...

> So it works, but what are the units on this `clock()` function? they are bizarre. Not milliseconds, or seconds. It's unclear Quoting https://man7.org/linux/man-pages/man3/clock.3.html: "The value returned is the CPU...

thanks, let's try it!

@olebole @GillesDuvert how to proceed here?

There are a few occurrences in the .pro files: https://github.com/search?q=repo%3Agnudatalanguage%2Fgdl+idlastro.gsfc&type=code