Merge pull request #41 from AMReX-Astro/bye-bye-summit

Summit is being decommissioned

sphinx_docs/source/olcf-compilers.rst

@@ -3,58 +3,6 @@
 Compiling at OLCF
 =================
 
-Summit
-------
-
-In order to compile you will need to swap the xl module with gcc (you need to use atleast gcc/7.4.0 due to C++17 support):
-
-.. prompt:: bash
-
-   module load gcc/10.2.0
-
-.. note::
-
-   You will need to load the same module you use for compiling in your
-   submissions script, otherwise the code won't find the shared
-   libraries at runtime.
-
-Then load CUDA:
-
-.. prompt:: bash
-
-  module load cuda/11.5.2
-
-.. note::
-
-   Presently you will see a warning when you load a CUDA 11 module, but the packages
-   should work fine.
-
-You also need to make sure you have the python module loaded:
-
-.. prompt:: bash
-
-  module load python
-
-Compile with ``USE_CUDA=TRUE`` (and ``COMP=gnu`` which is usually the default).
-Is important to setup ``USE_OMP=FALSE``, since the ``TRUE`` option is currently disallowed by Castro.
-An example compilation line is:
-
-.. prompt:: bash
-
-  make COMP=gnu USE_MPI=TRUE USE_CUDA=TRUE -j 4
-
-The recommended/tested version pairs are:
-
-  * ``gcc/10.2.0`` + ``cuda/11.5.2``
-
-.. note::
-
-   - OpenMP offloading to the GPU is controlled by
-     ``USE_OMP_OFFLOAD``, which will default to ``FALSE``, AMReX-Astro
-     doesn't use this feature.
-
-
-
 Frontier
 --------