Joe Schoonover

Now, only the mpas_dual test is failing. Is it possible that the dual grid creation is orienting vertices in the wrong direction (clockwise instead of counter-clockwise)??

> I'm also curious if you've had a chance to try out our recently optimized `Grid.get_faces_containing_point()` functionality. I haven't been following to closely on the spatial hashing work, but I'm...

Would you want me to attempt to fix the issue in this PR or should we resolve that separately ? I suppose we'd need a different test to check orientation...

@aaronzedwick - Got it. I'll see if I can come up with something that doesn't involve the spatial hashing where this might also crop up.

I did some testing and found that the failures in spatial hash query on the MPAS dual grid failures are not related to vertex ordering within elements. I did the...

Hey @cgmb, @domcharrier, @ppanchad-amd can we get a review on this ? The `.f` extension, which has historically been used in hipfort is used to identify Fortran77 Fixed format source...

@JorgeG94 - The CI builds apparently are failing with ``` -- The Fortran compiler identification is GNU 11.4.0 -- The C compiler identification is GNU 11.4.0 -- The CXX compiler...

> This seems reasonable to me, although I think there will be some conflict in the CMake with #222. Agreed. I guess we need to decide which branch to merge...

@JorgeG94 - It looks like the azuredevops CI definition for what hipfort actually runs is here : https://github.com/ROCm/ROCm/blob/develop/.azuredevops/components/hipfort.yml in the rocm/rocm repository

@JorgeG94 > I still need to use cmake -DHIPFORT_COMPILER=/usr/bin/gfortran -DHIPFORT_INSTALL_DIR=$HOME/install/hipfort ../ for the cmake to succeed, then everything fails at compile time with some files due to line truncation. What...