NEKO v0.9.1

.github/workflows/check_nvidia.yml

@@ -32,13 +32,13 @@ jobs:
               curl https://developer.download.nvidia.com/hpc-sdk/ubuntu/DEB-GPG-KEY-NVIDIA-HPC-SDK | sudo gpg --dearmor -o /usr/share/keyrings/nvidia-hpcsdk-archive-keyring.gpg
               echo 'deb [signed-by=/usr/share/keyrings/nvidia-hpcsdk-archive-keyring.gpg] https://developer.download.nvidia.com/hpc-sdk/ubuntu/amd64 /' | sudo tee /etc/apt/sources.list.d/nvhpc.list
               sudo apt-get update -y
-              sudo apt-get install -y nvhpc-24-7
+              sudo apt-get install -y nvhpc-24-11
               NVARCH=`uname -s`_`uname -m`
               export NVARCH
               NVCOMPILERS=/opt/nvidia/hpc_sdk
               export NVCOMPILERS
-              PATH=$NVCOMPILERS/$NVARCH/24.7/compilers/bin:$PATH; export PATH
-              export PATH=$NVCOMPILERS/$NVARCH/24.7/comm_libs/mpi/bin:$PATH
+              PATH=$NVCOMPILERS/$NVARCH/24.11/compilers/bin:$PATH; export PATH
+              export PATH=$NVCOMPILERS/$NVARCH/24.11/comm_libs/mpi/bin:$PATH
               printenv >> $GITHUB_ENV
               echo "os-version=$(lsb_release -ds | tr " " -)" >> $GITHUB_OUTPUT
 
@@ -98,7 +98,7 @@ jobs:
         run: |
           git apply patches/nvhpc_bge.patch
           ./regen.sh
-          ./configure FC=${FC} FCFLAGS="-O3" --enable-real=${RP} --with-cuda=/opt/nvidia/hpc_sdk/Linux_x86_64/24.7/cuda/
+          ./configure FC=${FC} FCFLAGS="-O3" --enable-real=${RP} --with-cuda=/opt/nvidia/hpc_sdk/Linux_x86_64/24.11/cuda/
           make
 
       - name: Dist (CPU backend)
@@ -122,5 +122,5 @@ jobs:
           tar xf neko-*.tar.gz -C releng
           cd releng/neko-*
           patch -u src/common/signal.f90 -i patches/nvhpc_bge.patch
-          ./configure FC=${FC} FCFLAGS="-O3" --enable-real=${RP} --with-cuda=/opt/nvidia/hpc_sdk/Linux_x86_64/24.7/cuda/
+          ./configure FC=${FC} FCFLAGS="-O3" --enable-real=${RP} --with-cuda=/opt/nvidia/hpc_sdk/Linux_x86_64/24.11/cuda/
           make -j $(nproc)

configure.ac

@@ -1,5 +1,5 @@
 AC_PREREQ([2.69])
-AC_INIT([neko],[0.9.0])
+AC_INIT([neko],[0.9.1])
 AM_INIT_AUTOMAKE([foreign subdir-objects])
 AM_MAINTAINER_MODE
 AC_CONFIG_MACRO_DIR([m4])

src/gs/gather_scatter.f90

@@ -450,6 +450,10 @@ subroutine gs_init_mapping(gs)
           end if
        end if
     end do
+
+    ! Clear local dofmap table
+    call dm%clear()
+
     if (lz .gt. 1) then
        !
        ! Setup mapping for dofs on edges
@@ -654,6 +658,10 @@ subroutine gs_init_mapping(gs)
           end if
        end do
     end if
+
+    ! Clear local dofmap table
+    call dm%clear()
+
     !
     ! Setup mapping for dofs on facets
     !

src/krylov/krylov.f90

@@ -189,7 +189,7 @@ end subroutine ksp_t_free
      !! @param monitor Enable/disable monitoring, optional.
      module subroutine krylov_solver_factory(object, n, type_name, &
           max_iter, abstol, M, monitor)
-       class(ksp_t), allocatable, target, intent(inout) :: object
+       class(ksp_t), allocatable, intent(inout) :: object
        integer, intent(in), value :: n
        character(len=*), intent(in) :: type_name
        integer, intent(in) :: max_iter

src/krylov/krylov_fctry.f90

@@ -76,7 +76,7 @@
   !! @param monitor Enable/disable residual history, optional.
   module subroutine krylov_solver_factory(object, n, type_name, &
        max_iter, abstol, M, monitor)
-    class(ksp_t), allocatable, target, intent(inout) :: object
+    class(ksp_t), allocatable, intent(inout) :: object
     integer, intent(in), value :: n
     character(len=*), intent(in) :: type_name
     integer, intent(in) :: max_iter

src/krylov/precon.f90

@@ -68,7 +68,7 @@ end subroutine pc_update
   interface
      !> Create a preconditioner
      module subroutine precon_factory(pc, type_name)
-       class(pc_t), target, allocatable, intent(inout) :: pc
+       class(pc_t),  allocatable, intent(inout) :: pc
        character(len=*), intent(in) :: type_name
      end subroutine precon_factory
 

src/krylov/precon_fctry.f90

@@ -51,7 +51,7 @@
 
   !> Create a preconditioner
   module subroutine precon_factory(pc, type_name)
-    class(pc_t), target, allocatable, intent(inout) :: pc
+    class(pc_t), allocatable, intent(inout) :: pc
     character(len=*), intent(in) :: type_name
     character(len=:), allocatable :: type_string
 

src/les/les_model.f90

@@ -114,7 +114,7 @@ end subroutine les_model_free
      !! @param coef SEM coefficients.
      !! @param json A dictionary with parameters.
      module subroutine les_model_factory(object, type_name, dofmap, coef, json)
-       class(les_model_t), allocatable, target, intent(inout) :: object
+       class(les_model_t), allocatable, intent(inout) :: object
        character(len=*), intent(in) :: type_name
        type(dofmap_t), intent(in) :: dofmap
        type(coef_t), intent(in) :: coef

src/les/les_model_fctry.f90

@@ -53,7 +53,7 @@
   !! @param coef SEM coefficients.
   !! @param json A dictionary with parameters.
   module subroutine les_model_factory(object, type_name, dofmap, coef, json)
-    class(les_model_t), allocatable, target, intent(inout) :: object
+    class(les_model_t), allocatable, intent(inout) :: object
     character(len=*), intent(in) :: type_name
     type(dofmap_t), intent(in) :: dofmap
     type(coef_t), intent(in) :: coef

src/sem/dofmap.f90

@@ -240,7 +240,7 @@ subroutine dofmap_number_edges(this)
     num_dofs_edges(1) =  int(Xh%lx - 2, i8)
     num_dofs_edges(2) =  int(Xh%ly - 2, i8)
     num_dofs_edges(3) =  int(Xh%lz - 2, i8)
-    edge_offset = int(msh%glb_mpts, i8) + int(1, 4)
+    edge_offset = int(msh%glb_mpts, i8) + int(1, i8)
 
     do i = 1, msh%nelv
 
@@ -380,7 +380,7 @@ subroutine dofmap_number_edges(this)
              end do
           end if
 
-          call ep%edge_id(edge, i8)
+          call ep%edge_id(edge, 8)
           shared_dof = msh%is_shared(edge)
           global_id = msh%get_global(edge)
           edge_id = edge_offset + int((global_id - 1), i8) * num_dofs_edges(2)

src/simulation_components/derivative.f90

@@ -149,7 +149,7 @@ subroutine derivative_compute(this, t, tstep)
     real(kind=rp), intent(in) :: t
     integer, intent(in) :: tstep
 
-    call dudxyz(this%du%x, this%u%x, this%dr, this%dr, this%dr,&
+    call dudxyz(this%du%x, this%u%x, this%dr, this%ds, this%dt,&
                 this%case%fluid%c_Xh)
   end subroutine derivative_compute
 

src/wall_models/wall_model.f90

@@ -156,7 +156,7 @@ end subroutine wall_model_free
      !! @param json A dictionary with parameters.
      module subroutine wall_model_factory(object, coef, msk, facet, nu, &
           json)
-       class(wall_model_t), allocatable, target, intent(inout) :: object
+       class(wall_model_t), allocatable, intent(inout) :: object
        type(coef_t), intent(in) :: coef
        integer, intent(in) :: msk(:)
        integer, intent(in) :: facet(:)

src/wall_models/wall_model_fctry.f90

@@ -55,7 +55,7 @@
   !! @param json A dictionary with parameters.
   module subroutine wall_model_factory(object, coef, msk, facet, nu, &
        json)
-    class(wall_model_t), allocatable, target, intent(inout) :: object
+    class(wall_model_t), allocatable, intent(inout) :: object
     type(coef_t), intent(in) :: coef
     integer, intent(in) :: msk(:)
     integer, intent(in) :: facet(:)