Shoshana Jakobovits

From @dev-zero: > CP2K and DBCSR are still in a co-development state (see timings between rc-releases of DBCSR and updates of the submodule), except for the accelerator part obviously. At...

> 3. Cut our losses and move DBCSR back into the CP2K repro. I am very strongly against this, for the same reasons as @alazzaro. Let's not go back. >...

> CP2K will build DBCSR via the toolchain I'm sorry, I have a very naive question: given that [cp2k is available in Spack](https://spack.readthedocs.io/en/latest/package_list.html#cp2k), - in cp2k's installation instructions, why aren't...

@dev-zero Thanks a lot for all your work on Spack! > CP2K will build DBCSR via the toolchain Given that we all seem to believe that Spack is the way...

I very much agree. Are there any current obstacles that need to be lifted in order to get a CI for GPU and what needs to be done?

Great. If there's something I can help with, do let me know. Good luck with the approaching deadlines :)

Linking to PR #241

Sorry for chiming in a little late. > At this point, I wonder if this is something new... we never had a similar problem in the past (even if it...

In the meantime, [hiprtc](https://github.com/ROCm-Developer-Tools/HIP/blob/master/src/hiprtc.cpp) has been added to HIP, thus making it possible to compile kernels at runtime on AMD GPUs 🎉 Unfortunately however, it isn't possible to compile any...