This PR fixes issues #1054 and #1060. Looks like coeff needs to be a ScalarT instead of a double in order to get the correct derivative values in DistParamDeriv local responses.
This also restores the uvm-free kokkos implementation of ResponseGLFlux.
This PR fixes issues #1054 and #1060. Looks like
coeff
needs to be a ScalarT instead of a double in order to get the correct derivative values in DistParamDeriv local responses.This also restores the uvm-free kokkos implementation of ResponseGLFlux.