Merge branch 'ndk/machinefiles/pm-gpu-Kokkos_ENABLE_IMPL_CUDA_MALLOC_…

…ASYNC=OFF' into next (PR #6423)

After #6101 which brings in kokkos 4.2, we see runtime error with a test like:
ERP_Ln9.ne4pg2_ne4pg2.F2010-SCREAMv1.pm-gpu_gnugpu

hits runtime error like:

0: (GTL DEBUG: 0) cuIpcGetMemHandle: invalid argument, CUDA_ERROR_INVALID_VALUE, line no 97
unfortunately, the tests hitting this error are also hanging...

A fix that seems to work is to add this build flag:
-DKokkos_ENABLE_IMPL_CUDA_MALLOC_ASYNC=OFF

The fix is merged in another PR for pm-gpu, so this PR just makes same change to muller-gpu.

Fixes #6422

cime_config/machines/cmake_macros/gnugpu_muller-gpu.cmake

@@ -6,7 +6,8 @@ if (COMP_NAME STREQUAL gptl)
 endif()
 string(APPEND CPPDEFS " -DTHRUST_IGNORE_CUB_VERSION_CHECK")
 string(APPEND CMAKE_CUDA_FLAGS " -ccbin CC -O2 -arch sm_80 --use_fast_math")
-string(APPEND KOKKOS_OPTIONS " -DKokkos_ARCH_AMPERE80=On -DKokkos_ENABLE_CUDA=On -DKokkos_ENABLE_CUDA_LAMBDA=On -DKokkos_ENABLE_SERIAL=ON -DKokkos_ENABLE_OPENMP=Off")
+string(APPEND KOKKOS_OPTIONS " -DKokkos_ARCH_AMPERE80=On -DKokkos_ENABLE_CUDA=On -DKokkos_ENABLE_CUDA_LAMBDA=On -DKokkos_ENABLE_SERIAL=ON -DKokkos_ENABLE_OPENMP=Off -DKokkos_ENABLE_IMPL_CUDA_MALLOC_ASYNC=OFF")
+set(CMAKE_CUDA_ARCHITECTURES "80")
 string(APPEND CMAKE_C_FLAGS_RELEASE " -O2")
 string(APPEND CMAKE_Fortran_FLAGS_RELEASE " -O2")
 set(MPICC "cc")