Tom Epperly
Tom Epperly
I guess the same applies to cub and rocprim.
Our index types are all defined using RAJA_INDEX_VALUE(TimeInd, "Time Index"), RAJA_INDEX_VALUE(YInd, "Y Spatial Index), etc. Underneath, they are all RAJA::Index_type. We could use something like a int3 or std::tuple.
It would be nice to know on what time scale a feature like this might appear (days, weeks, months, or years). If it's days or weeks, I would probably wait....
I also think this feature would be useful. It would be nice to be able to preserve comments the user put into a file when rewriting the file (as stated...
Okay to do option 1, I think I would need to follow a process like this: 1. Run my code on one MPI rank and save the result to a...
I think they plan to work on this. On Wed, May 25, 2022, 3:29 PM stale[bot] ***@***.***> wrote: > ⚠️ This issue or PR has been automatically marked as stale...
Fedora has gcc 11.3.1, and maybe it's worse? :-) Tom On Tue, Nov 8, 2022 at 8:56 AM Rich Hornung ***@***.***> wrote: > @tepperly If you grab the latest RAJA...
This change in `mfem` requires this patch to serac: ```diff diff --git a/src/serac/numerics/equation_solver.hpp b/src/serac/numerics/equation_solver.hpp index dfd2d1cc..e86f5d0d 100644 --- a/src/serac/numerics/equation_solver.hpp +++ b/src/serac/numerics/equation_solver.hpp @@ -203,11 +203,11 @@ public: * @param[in] comm The...
The change to mfem's Strumpack wrapper plugs bigger leaks than #1032 . Without the mfem change, I can only run