Kripke status

[pearce8]

Components

1. Source code, Build system: CMake
2. Source code released
3. Spack package.py
4. Ramble application.py
5. Benchpark experiment.py
6. +cuda experiment.py
7. +rocm experiment.py
8. Benchmark dependencies: RAJA, Umpire, Chai (consider removing)
9. small test size problem for CI
10. Scaling (Weak, Strong, or both)
11. Large scale problem size defined
12. CI running on CPUs/CDash
13. CI running on NVIDIA GPUs/CDash
14. CI running on AMD GPUs/CDash
15. +single_memory experiment.py Umpire/CHAI changes  Kripke#70
16. +caliper experiment.py
17. CPU-only performance analysis
18. CUDA performance analysis
19. ROCm performance analysis
20. Documentation (build, run instructions, performance demo)
21. Add Umpire and force Chai to allocate before the main loop. Fixed in Umpire device pool for Kripke Kripke#65
22. Robert working on resolving race condition in LTimes DRAFT: Fix for Race Condition in Times Kripke#71
23. Problem size: 8x8 solve on each thread generates register pressure. How to pick the size based on memory?