-
Currently there are many empty `OMPGPU` regions in develop which we need to fill in. When merging the optimised branch, we prioritised OpenACC offloading, and now that that is relatively mature, we sh…
-
Dear Dr. Ye Yuan,
Is the sediment model of FUNWAVE-GPU updated version or not, please?
Your kind reply is highly appreciated.
-
### Untested code
The untested code is:
https://github.com/pyccel/pyccel/blob/b10051d8bfd555da911f1489f23bc358abe252ce/pyccel/parser/syntax/openacc.py
https://github.com/pyccel/pyccel/blob/b10051…
-
The OpenACC implementation requires GCC 6+ to compile, but then only with a patch.
Please rename the patch to remove the `.txt` suffix, which Github forced me to add.
[0001-use-__restrict__-instea…
-
As I did with CUDA(#1028) and I tried to do with Arrayfire(#1049) and [pytorch](https://github.com/Iximiel/plumed2/tree/torchCoord/plugins/torchCoord), I tried to [rewrite](https://github.com/Iximiel/…
-
-
Hi,
I am wondering if a GPU version of VASP can be compiled and if compilation instructions can be provided. I have tried to compile it using the existing nvompi/2022.7 module, but one of the main …
-
Hi, I am not sure it can be run project with OpenACC+fortran.
With nvfortran command. For example,
On Saga, load the NVHPC/20.7 module and compile with nvfortran as follows:
$ module load NVHPC/…
-
nvidia optimised this benchmark and wrote a report explaining what they did. We should capture their optimisations in a manual implementation and test them out. We should also compare these versions w…
-
> It _should_ be relatively straightforward to extend the frontend to handle OpenACC directives.
In fact, a first step would be simply to put such directives into CodeBlocks.
_Orig…