DKS Musrfit module contains functions to perform chi squared and max-log likelihood calculations as well as provides the mechanism to create and compile the GPU code at runtime. Musrfit targets GPUs to accelerate the calculations, therefore DKS contains kernel implementations using CUDA and OpenCL to allow targeting of Nvidia and AMD devices. The OpenCL implementation also allows using the CPU and potentially Intel MIC devices for acceleration, although the code is optimized specifically for GPU architectures.
Full list of DKSBaseMuSR module functions are available in DKS documentation. The full description of DKSMusrfit module and its interaction with musrfit can be found in Thesis Chapter 5.