From b153d6f6648cd188c1a2698c730a76300b3e62c2 Mon Sep 17 00:00:00 2001
From: Arturo Vargas <vargas45@llnl.gov>
Date: Thu, 19 Sep 2024 10:25:40 -0700
Subject: [PATCH 1/8] attempt at fixing build errors

---
 benchmark/CMakeLists.txt             |   2 +
 benchmark/ltimes.cpp                 | 220 +++++++++++++--------------
 scripts/lc-builds/blueos_nvcc_gcc.sh |   1 +
 scripts/lc-builds/toss4_amdclang.sh  |   1 +
 4 files changed, 114 insertions(+), 110 deletions(-)

diff --git a/benchmark/CMakeLists.txt b/benchmark/CMakeLists.txt
index 8ddeba454d..15d919c566 100644
--- a/benchmark/CMakeLists.txt
+++ b/benchmark/CMakeLists.txt
@@ -11,9 +11,11 @@ if (RAJA_ENABLE_CUDA)
     SOURCES host-device-lambda-benchmark.cpp)
 endif()
 
+if (RAJA_ENABLE_DESUL)
 raja_add_benchmark(
   NAME benchmark-atomic
   SOURCES benchmark-atomic.cpp)
+endif()
 
 raja_add_benchmark(
   NAME ltimes
diff --git a/benchmark/ltimes.cpp b/benchmark/ltimes.cpp
index b2fa413b8e..c85826eff8 100644
--- a/benchmark/ltimes.cpp
+++ b/benchmark/ltimes.cpp
@@ -88,18 +88,18 @@ extern "C" {
  *      RAJA 'statement' concepts
  *
  *  Note that calls to the checkResult() method after each variant is run
- *  are turned off so the example code runs much faster. If you want 
+ *  are turned off so the example code runs much faster. If you want
  *  to verify the results are correct, define the 'DEBUG_LTIMES' macro
  *  below or turn on checking for individual variants.
  */
 
 
 
-
 using namespace RAJA;
+using namespace RAJA::expt;
 
 //
-// Index value types for strongly-typed indices must be defined outside 
+// Index value types for strongly-typed indices must be defined outside
 // function scope for RAJA CUDA variants to work.
 //
 // These types provide strongly-typed index values so if something is wrong
@@ -116,7 +116,7 @@ RAJA_INDEX_VALUE_T(IZ, int, "IZ");
 //
 template <typename PHIVIEW_T, typename LVIEW_T, typename PSIVIEW_T>
 void checkResult(PHIVIEW_T& phi, LVIEW_T& L, PSIVIEW_T& psi,
-                 const int num_m, 
+                 const int num_m,
                  const int num_d,
                  const int num_g,
                  const int num_z);
@@ -231,14 +231,14 @@ int main(int RAJA_UNUSED_ARG(argc), char **RAJA_UNUSED_ARG(argv[]))
 
   //
   // View types and Views/Layouts for indexing into arrays
-  // 
-  // L(m, d) : 1 -> d is stride-1 dimension 
+  //
+  // L(m, d) : 1 -> d is stride-1 dimension
   using LView = TypedView<double, Layout<2, int, 0>, IM, ID>;
 
-  // psi(d, g, z) : 2 -> z is stride-1 dimension 
+  // psi(d, g, z) : 2 -> z is stride-1 dimension
   using PsiView = TypedView<double, Layout<3, int, 0>, ID, IG, IZ>;
 
-  // phi(m, g, z) : 2 -> z is stride-1 dimension 
+  // phi(m, g, z) : 2 -> z is stride-1 dimension
   using PhiView = TypedView<double, Layout<3, int, 0>, IM, IG, IZ>;
 
   std::array<RAJA::idx_t, 2> L_perm {{1, 0}};
@@ -255,7 +255,7 @@ int main(int RAJA_UNUSED_ARG(argc), char **RAJA_UNUSED_ARG(argv[]))
 
 
   RAJA::Timer timer;
-  timer.start(); 
+  timer.start();
 
   for (int iter = 0;iter < num_iter;++ iter)
     for (IG g(0); g < num_g; ++g) {
@@ -268,7 +268,7 @@ int main(int RAJA_UNUSED_ARG(argc), char **RAJA_UNUSED_ARG(argv[]))
     }
   }
 
-  timer.stop(); 
+  timer.stop();
   double t = timer.elapsed();
   double gflop_rate = total_flops / t / 1.0e9;
   std::cout << "  C-version of LTimes run time (with Views) (sec.): "
@@ -291,14 +291,14 @@ int main(int RAJA_UNUSED_ARG(argc), char **RAJA_UNUSED_ARG(argv[]))
 
   //
   // View types and Views/Layouts for indexing into arrays
-  // 
-  // L(m, d) : 1 -> d is stride-1 dimension 
+  //
+  // L(m, d) : 1 -> d is stride-1 dimension
   using LView = TypedView<double, Layout<2, int, 0>, IM, ID>;
 
-  // psi(d, g, z) : 2 -> z is stride-1 dimension 
+  // psi(d, g, z) : 2 -> z is stride-1 dimension
   using PsiView = TypedView<double, Layout<3, int, 0>, ID, IG, IZ>;
 
-  // phi(m, g, z) : 2 -> z is stride-1 dimension 
+  // phi(m, g, z) : 2 -> z is stride-1 dimension
   using PhiView = TypedView<double, Layout<3, int, 0>, IM, IG, IZ>;
 
   std::array<RAJA::idx_t, 2> L_perm {{1, 0}};
@@ -313,11 +313,11 @@ int main(int RAJA_UNUSED_ARG(argc), char **RAJA_UNUSED_ARG(argv[]))
   PhiView phi(phi_data,
               RAJA::make_permuted_layout({{num_m, num_g, num_z}}, phi_perm));
 
-  using EXECPOL = 
+  using EXECPOL =
     RAJA::KernelPolicy<
-      statement::For<2, loop_exec,  // g
-        statement::For<3, loop_exec,  // z
-         statement::For<0, loop_exec,  // m
+      statement::For<2, seq_exec,  // g
+        statement::For<3, seq_exec,  // z
+         statement::For<0, seq_exec,  // m
            statement::For<1, simd_exec,  // d
                statement::Lambda<0>
              >
@@ -388,9 +388,9 @@ int main(int RAJA_UNUSED_ARG(argc), char **RAJA_UNUSED_ARG(argv[]))
 
   using EXECPOL =
     RAJA::KernelPolicy<
-      statement::For<2, loop_exec,  // g
-        statement::For<3, loop_exec,  // z
-           statement::For<0, loop_exec,  // m
+      statement::For<2, seq_exec,  // g
+        statement::For<3, seq_exec,  // z
+           statement::For<0, seq_exec,  // m
              statement::For<1, simd_exec,  // d
                statement::Lambda<0, Segs<0, 1, 2, 3>>
              >
@@ -462,10 +462,10 @@ int main(int RAJA_UNUSED_ARG(argc), char **RAJA_UNUSED_ARG(argv[]))
 
 
   using pol_launch = RAJA::LaunchPolicy<RAJA::seq_launch_t>;
-  using pol_g = RAJA::LoopPolicy<RAJA::loop_exec>;
-  using pol_z = RAJA::LoopPolicy<RAJA::loop_exec>;
-  using pol_m = RAJA::LoopPolicy<RAJA::loop_exec>;
-  using pol_d = RAJA::LoopPolicy<RAJA::loop_exec>;
+  using pol_g = RAJA::LoopPolicy<RAJA::seq_exec>;
+  using pol_z = RAJA::LoopPolicy<RAJA::seq_exec>;
+  using pol_m = RAJA::LoopPolicy<RAJA::seq_exec>;
+  using pol_d = RAJA::LoopPolicy<RAJA::seq_exec>;
 
 
 
@@ -535,14 +535,14 @@ int main(int RAJA_UNUSED_ARG(argc), char **RAJA_UNUSED_ARG(argv[]))
   PhiView phi(phi_data,
               RAJA::make_permuted_layout({{num_m, num_g, num_z}}, phi_perm));
 
-  using vector_t = RAJA::VectorRegister<double>;
-  using VecIZ = RAJA::VectorIndex<IZ, vector_t>;
+  using vector_t = RAJA::expt::VectorRegister<double>;
+  using VecIZ = RAJA::expt::VectorIndex<IZ, vector_t>;
 
   using EXECPOL =
     RAJA::KernelPolicy<
-      statement::For<2, loop_exec,  // g
-        statement::For<0, loop_exec,  // m
-          statement::For<1, loop_exec,  // d
+      statement::For<2, seq_exec,  // g
+        statement::For<0, seq_exec,  // m
+          statement::For<1, seq_exec,  // d
 
              statement::Lambda<0>
            >
@@ -552,7 +552,7 @@ int main(int RAJA_UNUSED_ARG(argc), char **RAJA_UNUSED_ARG(argv[]))
 
 
 #ifdef RAJA_ENABLE_VECTOR_STATS
-  RAJA::tensor_stats::resetVectorStats();
+  RAJA::expt::tensor_stats::resetVectorStats();
 #endif
 
 
@@ -622,9 +622,9 @@ int main(int RAJA_UNUSED_ARG(argc), char **RAJA_UNUSED_ARG(argv[]))
   PhiView phi(phi_data,
               RAJA::make_permuted_layout({{num_m, num_g, num_z}}, phi_perm));
 
-  using matrix_t = RAJA::SquareMatrixRegister<double, ColMajorLayout, RAJA::scalar_register>;
-  //using matrix_t = RAJA::SquareMatrixRegister<double, RowMajorLayout>;
-//  using matrix_t = RAJA::RectMatrixRegister<double, RAJA::ColMajorLayout, 8,8>;
+  using matrix_t = RAJA::expt::SquareMatrixRegister<double, ColMajorLayout, RAJA::expt::scalar_register>;
+  //using matrix_t = RAJA::expt::SquareMatrixRegister<double, RowMajorLayout>;
+//  using matrix_t = RAJA::expt::RectMatrixRegister<double, RAJA::ColMajorLayout, 8,8>;
 
 
 
@@ -633,9 +633,9 @@ int main(int RAJA_UNUSED_ARG(argc), char **RAJA_UNUSED_ARG(argv[]))
 
 	printf("Num registers/matrix = %d\n", (int)matrix_t::s_num_registers);
 
-  using RowM = RAJA::RowIndex<IM, matrix_t>;
-  using ColD = RAJA::ColIndex<ID, matrix_t>;
-  using ColZ = RAJA::ColIndex<IZ, matrix_t>;
+  using RowM = RAJA::expt::RowIndex<IM, matrix_t>;
+  using ColD = RAJA::expt::ColIndex<ID, matrix_t>;
+  using ColZ = RAJA::expt::ColIndex<IZ, matrix_t>;
 
 
 #ifdef RAJA_ENABLE_VECTOR_STATS
@@ -648,7 +648,7 @@ int main(int RAJA_UNUSED_ARG(argc), char **RAJA_UNUSED_ARG(argv[]))
 
   for (int iter = 0;iter < num_iter;++ iter){
 
-  RAJA::forall<loop_exec>(RAJA::TypedRangeSegment<IG>(0, num_g),
+  RAJA::forall<seq_exec>(RAJA::TypedRangeSegment<IG>(0, num_g),
     [=](IG g)
     {
 
@@ -718,19 +718,19 @@ int main(int RAJA_UNUSED_ARG(argc), char **RAJA_UNUSED_ARG(argv[]))
   PhiView phi(phi_data,
               RAJA::make_permuted_layout({{num_m, num_g, num_z}}, phi_perm));
 
-    using matrix_t = RAJA::SquareMatrixRegister<double, RowMajorLayout>;
+  using matrix_t = RAJA::expt::SquareMatrixRegister<double, RowMajorLayout>;
 
 
     std::cout << "matrix size: " << matrix_t::s_dim_elem(0) <<
         "x" << matrix_t::s_dim_elem(1) << std::endl;
 
-    using RowM = RAJA::RowIndex<IM, matrix_t>;
-    using ColD = RAJA::ColIndex<ID, matrix_t>;
-    using ColZ = RAJA::ColIndex<IZ, matrix_t>;
+    using RowM = RAJA::expt::RowIndex<IM, matrix_t>;
+    using ColD = RAJA::expt::ColIndex<ID, matrix_t>;
+    using ColZ = RAJA::expt::ColIndex<IZ, matrix_t>;
 
 
   #ifdef RAJA_ENABLE_VECTOR_STATS
-    RAJA::tensor_stats::resetVectorStats();
+    RAJA::expt::tensor_stats::resetVectorStats();
   #endif
 
     RAJA::Timer timer;
@@ -738,7 +738,7 @@ int main(int RAJA_UNUSED_ARG(argc), char **RAJA_UNUSED_ARG(argv[]))
 
     for (int iter = 0;iter < num_iter;++ iter){
 
-    RAJA::forall<loop_exec>(RAJA::TypedRangeSegment<IG>(0, num_g),
+    RAJA::forall<seq_exec>(RAJA::TypedRangeSegment<IG>(0, num_g),
       [=](IG g)
       {
 
@@ -820,13 +820,13 @@ int main(int RAJA_UNUSED_ARG(argc), char **RAJA_UNUSED_ARG(argv[]))
       statement::InitLocalMem<RAJA::cpu_tile_mem, RAJA::ParamList<0,1,2>,
 
       // Tile outer m,d loops
-      statement::Tile<0, tile_fixed<tile_m>, loop_exec,  // m
-        statement::Tile<1, tile_fixed<tile_d>, loop_exec,  // d
+      statement::Tile<0, tile_fixed<tile_m>, seq_exec,  // m
+        statement::Tile<1, tile_fixed<tile_d>, seq_exec,  // d
 
 
             // Load L(m,d) for m,d tile into shmem
-            statement::For<0, loop_exec,  // m
-              statement::For<1, loop_exec,  // d
+            statement::For<0, seq_exec,  // m
+              statement::For<1, seq_exec,  // d
                 statement::Lambda<0, Segs<0, 1>,
                                      Params<0>,
                                      Offsets<0, 1>>
@@ -834,13 +834,13 @@ int main(int RAJA_UNUSED_ARG(argc), char **RAJA_UNUSED_ARG(argv[]))
             >,
 
             // Run inner g, z loops with z loop tiled
-            statement::For<2, loop_exec,  // g
-              statement::Tile<3, tile_fixed<tile_z>, loop_exec,  // z
+            statement::For<2, seq_exec,  // g
+              statement::Tile<3, tile_fixed<tile_z>, seq_exec,  // z
 
 
                   // Load psi into shmem
-                  statement::For<1, loop_exec,  // d
-                    statement::For<3, loop_exec,  // z
+                  statement::For<1, seq_exec,  // d
+                    statement::For<3, seq_exec,  // z
                       statement::Lambda<1, Segs<1, 2, 3>,
                                            Params<1>,
                                            Offsets<1, 2, 3>>
@@ -848,25 +848,25 @@ int main(int RAJA_UNUSED_ARG(argc), char **RAJA_UNUSED_ARG(argv[]))
                   >,
 
                   // Compute phi
-                  statement::For<0, loop_exec,  // m
+                  statement::For<0, seq_exec,  // m
 
                     // Load phi into shmem
-                    statement::For<3, loop_exec,  // z
+                    statement::For<3, seq_exec,  // z
                       statement::Lambda<2, Segs<0, 2, 3>,
                                            Params<2>,
                                            Offsets<0, 2, 3>>
                     >,
 
                     // Compute phi in shmem
-                    statement::For<1, loop_exec,  // d
-                      statement::For<3, loop_exec,  // z
+                    statement::For<1, seq_exec,  // d
+                      statement::For<3, seq_exec,  // z
                         statement::Lambda<3, Params<0, 1, 2>,
                                              Offsets<0, 1, 2, 3>>
                       >
                     >,
 
                     // Store phi
-                    statement:: For<3, loop_exec,  // z
+                    statement:: For<3, seq_exec,  // z
                       statement::Lambda<4, Segs<0, 2, 3>,
                                            Params<2>,
                                            Offsets<0, 2, 3>>
@@ -992,14 +992,14 @@ int main(int RAJA_UNUSED_ARG(argc), char **RAJA_UNUSED_ARG(argv[]))
 
   //
   // View types and Views/Layouts for indexing into arrays
-  // 
-  // L(m, d) : 1 -> d is stride-1 dimension 
+  //
+  // L(m, d) : 1 -> d is stride-1 dimension
   using LView = TypedView<double, Layout<2, int, 1>, IM, ID>;
 
-  // psi(d, g, z) : 2 -> z is stride-1 dimension 
+  // psi(d, g, z) : 2 -> z is stride-1 dimension
   using PsiView = TypedView<double, Layout<3, int, 2>, ID, IG, IZ>;
 
-  // phi(m, g, z) : 2 -> z is stride-1 dimension 
+  // phi(m, g, z) : 2 -> z is stride-1 dimension
   using PhiView = TypedView<double, Layout<3, int, 2>, IM, IG, IZ>;
 
   std::array<RAJA::idx_t, 2> L_perm {{0, 1}};
@@ -1019,9 +1019,9 @@ int main(int RAJA_UNUSED_ARG(argc), char **RAJA_UNUSED_ARG(argv[]))
   using EXECPOL =
     RAJA::KernelPolicy<
       statement::For<0, omp_parallel_for_exec,  // m
-        statement::For<1, loop_exec,  // d 
-          statement::For<2, loop_exec,  // g 
-            statement::For<3, simd_exec,  // z 
+        statement::For<1, seq_exec,  // d
+          statement::For<2, seq_exec,  // g
+            statement::For<3, simd_exec,  // z
               statement::Lambda<0>
             >
           >
@@ -1037,7 +1037,7 @@ int main(int RAJA_UNUSED_ARG(argc), char **RAJA_UNUSED_ARG(argv[]))
     RAJA::KernelPolicy<
       statement::Collapse<omp_parallel_collapse_exec,
                           RAJA::ArgList<0, 2, 3>,   // m, g, z
-        statement::For<1, loop_exec,  // d 
+        statement::For<1, seq_exec,  // d
           statement::Lambda<0>
         >
       >
@@ -1096,14 +1096,14 @@ int main(int RAJA_UNUSED_ARG(argc), char **RAJA_UNUSED_ARG(argv[]))
 
   //
   // View types and Views/Layouts for indexing into arrays
-  // 
-  // L(m, d) : 1 -> d is stride-1 dimension 
+  //
+  // L(m, d) : 1 -> d is stride-1 dimension
   using LView = TypedView<double, Layout<2, int, 1>, IM, ID>;
 
-  // psi(d, g, z) : 2 -> z is stride-1 dimension 
+  // psi(d, g, z) : 2 -> z is stride-1 dimension
   using PsiView = TypedView<double, Layout<3, int, 2>, ID, IG, IZ>;
 
-  // phi(m, g, z) : 2 -> z is stride-1 dimension 
+  // phi(m, g, z) : 2 -> z is stride-1 dimension
   using PhiView = TypedView<double, Layout<3, int, 2>, IM, IG, IZ>;
 
   std::array<RAJA::idx_t, 2> L_perm {{0, 1}};
@@ -1120,7 +1120,7 @@ int main(int RAJA_UNUSED_ARG(argc), char **RAJA_UNUSED_ARG(argv[]))
 
   using EXECPOL =
     RAJA::KernelPolicy<
-      statement::CudaKernelAsync<    
+      statement::CudaKernelAsync<
         statement::For<0, cuda_block_x_loop,  // m
           statement::For<2, cuda_block_y_loop,  // g
             statement::For<3, cuda_thread_x_loop,  // z
@@ -1130,9 +1130,9 @@ int main(int RAJA_UNUSED_ARG(argc), char **RAJA_UNUSED_ARG(argv[]))
             >
           >
         >
-      >         
-    >;          
-                 
+      >
+    >;
+
   auto segments = RAJA::make_tuple(RAJA::TypedRangeSegment<IM>(0, num_m),
                                    RAJA::TypedRangeSegment<ID>(0, num_d),
                                    RAJA::TypedRangeSegment<IG>(0, num_g),
@@ -1165,9 +1165,9 @@ int main(int RAJA_UNUSED_ARG(argc), char **RAJA_UNUSED_ARG(argv[]))
   cudaErrchk( cudaFree( dphi_data ) );
 
   // Reset data in Views to CPU data
-  L.set_data(L_data); 
-  psi.set_data(psi_data); 
-  phi.set_data(phi_data); 
+  L.set_data(L_data);
+  psi.set_data(psi_data);
+  phi.set_data(phi_data);
 
 #if defined(DEBUG_LTIMES)
   checkResult(phi, L, psi, num_m, num_d, num_g, num_z);
@@ -1201,10 +1201,10 @@ int main(int RAJA_UNUSED_ARG(argc), char **RAJA_UNUSED_ARG(argv[]))
 
 
   using pol_launch = RAJA::LaunchPolicy<RAJA::seq_launch_t, RAJA::cuda_launch_t<true, 512> >;
-  using pol_g = RAJA::LoopPolicy<RAJA::loop_exec, cuda_block_x_loop>;
-  using pol_z = RAJA::LoopPolicy<RAJA::loop_exec, cuda_thread_y_loop>;
-  using pol_m = RAJA::LoopPolicy<RAJA::loop_exec, cuda_thread_x_loop>;
-  using pol_d = RAJA::LoopPolicy<RAJA::loop_exec, RAJA::loop_exec>;
+  using pol_g = RAJA::LoopPolicy<RAJA::seq_exec, cuda_block_x_loop>;
+  using pol_z = RAJA::LoopPolicy<RAJA::seq_exec, cuda_thread_y_loop>;
+  using pol_m = RAJA::LoopPolicy<RAJA::seq_exec, cuda_thread_x_loop>;
+  using pol_d = RAJA::LoopPolicy<RAJA::seq_exec, RAJA::seq_exec>;
 
 
   //
@@ -1325,22 +1325,22 @@ int main(int RAJA_UNUSED_ARG(argc), char **RAJA_UNUSED_ARG(argv[]))
 
   using matrix_layout = RowMajorLayout;
 
-  using L_matrix_host_t = RAJA::SquareMatrixRegister<double, matrix_layout>;
-  using L_matrix_device_t = RAJA::RectMatrixRegister<double, matrix_layout, 8, 4, RAJA::cuda_warp_register>;
+  using L_matrix_host_t = RAJA::expt::SquareMatrixRegister<double, matrix_layout>;
+  using L_matrix_device_t = RAJA::expt::RectMatrixRegister<double, matrix_layout, 8, 4, RAJA::expt::cuda_warp_register>;
   using L_matrix_hd_t = RAJA::LaunchPolicy<L_matrix_host_t, L_matrix_device_t>;
 
-  using phi_matrix_host_t = RAJA::SquareMatrixRegister<double, matrix_layout>;
-  using phi_matrix_device_t = RAJA::RectMatrixRegister<double, matrix_layout, 8, 8, RAJA::cuda_warp_register>;
+  using phi_matrix_host_t = RAJA::expt::SquareMatrixRegister<double, matrix_layout>;
+  using phi_matrix_device_t = RAJA::expt::RectMatrixRegister<double, matrix_layout, 8, 8, RAJA::expt::cuda_warp_register>;
   using phi_matrix_hd_t = RAJA::LaunchPolicy<L_matrix_host_t, phi_matrix_device_t>;
 
-  using psi_matrix_host_t = RAJA::SquareMatrixRegister<double, matrix_layout>;
-  using psi_matrix_device_t = RAJA::RectMatrixRegister<double, matrix_layout, 4, 8, RAJA::cuda_warp_register>;
+  using psi_matrix_host_t = RAJA::expt::SquareMatrixRegister<double, matrix_layout>;
+  using psi_matrix_device_t = RAJA::expt::RectMatrixRegister<double, matrix_layout, 4, 8, RAJA::expt::cuda_warp_register>;
   using psi_matrix_hd_t = RAJA::LaunchPolicy<L_matrix_host_t, psi_matrix_device_t>;
 
 
   using pol_launch = RAJA::LaunchPolicy<RAJA::seq_launch_t, RAJA::cuda_launch_t<true , 1024> >;
-  using pol_g = RAJA::LoopPolicy<RAJA::loop_exec, cuda_block_x_direct>;
-  using pol_z = RAJA::LoopPolicy<RAJA::loop_exec, cuda_thread_y_loop>;
+  using pol_g = RAJA::LoopPolicy<RAJA::seq_exec, cuda_block_x_direct>;
+  using pol_z = RAJA::LoopPolicy<RAJA::seq_exec, cuda_thread_y_loop>;
 
 
   //
@@ -1388,16 +1388,16 @@ int main(int RAJA_UNUSED_ARG(argc), char **RAJA_UNUSED_ARG(argv[]))
 
 
       using L_matrix_t = RAJA_GET_POLICY(L_matrix_hd_t);
-      using L_RowM = RAJA::RowIndex<IM, L_matrix_t>;
-      using L_ColD = RAJA::ColIndex<ID, L_matrix_t>;
+      using L_RowM = RAJA::expt::RowIndex<IM, L_matrix_t>;
+      using L_ColD = RAJA::expt::ColIndex<ID, L_matrix_t>;
 
       using psi_matrix_t = RAJA_GET_POLICY(psi_matrix_hd_t);
-      using psi_RowD = RAJA::RowIndex<ID, psi_matrix_t>;
-      using psi_ColZ = RAJA::ColIndex<IZ, psi_matrix_t>;
+      using psi_RowD = RAJA::expt::RowIndex<ID, psi_matrix_t>;
+      using psi_ColZ = RAJA::expt::ColIndex<IZ, psi_matrix_t>;
 
       using phi_matrix_t = RAJA_GET_POLICY(phi_matrix_hd_t);
-      using phi_RowM = RAJA::RowIndex<IM, phi_matrix_t>;
-      using phi_ColZ = RAJA::ColIndex<IZ, phi_matrix_t>;
+      using phi_RowM = RAJA::expt::RowIndex<IM, phi_matrix_t>;
+      using phi_ColZ = RAJA::expt::ColIndex<IZ, phi_matrix_t>;
 
 
       RAJA::loop<pol_g>(ctx, RAJA::TypedRangeSegment<IG>(0, num_g), [&](IG g){
@@ -1468,8 +1468,8 @@ int main(int RAJA_UNUSED_ARG(argc), char **RAJA_UNUSED_ARG(argv[]))
 
   //
   // View types and Views/Layouts for indexing into arrays
-  // 
-  // L(m, d) : 1 -> d is stride-1 dimension 
+  //
+  // L(m, d) : 1 -> d is stride-1 dimension
   using LView = TypedView<double, Layout<2, int, 1>, IM, ID>;
 
   // psi(d, g, z) : 2 -> z is stride-1 dimension
@@ -1530,11 +1530,11 @@ int main(int RAJA_UNUSED_ARG(argc), char **RAJA_UNUSED_ARG(argv[]))
     RAJA::KernelPolicy<
       statement::CudaKernelAsync<
         statement::InitLocalMem<cuda_shared_mem, ParamList<0,1>,
-          // Tile outer m,d loops 
+          // Tile outer m,d loops
           statement::Tile<0, tile_fixed<tile_m>, seq_exec,  // m
             statement::Tile<1, tile_fixed<tile_d>, seq_exec,  // d
 
-              // Load L for m,d tile into shmem 
+              // Load L for m,d tile into shmem
               statement::For<1, cuda_thread_x_loop,  // d
                 statement::For<0, cuda_thread_y_direct,   // m
                   statement::Lambda<0, Segs<0,1>, Params<0>, Offsets<0,1>>
@@ -1571,7 +1571,7 @@ int main(int RAJA_UNUSED_ARG(argc), char **RAJA_UNUSED_ARG(argv[]))
                       > // d
                     >  // m
                   >,  // z
-                  
+
                   // finish tile over directions
                   statement::CudaSyncThreads,
 
@@ -1582,7 +1582,7 @@ int main(int RAJA_UNUSED_ARG(argc), char **RAJA_UNUSED_ARG(argv[]))
                     >
                   >,
                   statement::CudaSyncThreads
-                
+
                 >  // Tile z
               >  // g
 
@@ -1594,7 +1594,7 @@ int main(int RAJA_UNUSED_ARG(argc), char **RAJA_UNUSED_ARG(argv[]))
     >;  // KernelPolicy
 
 
-  
+
 
 
   RAJA::Timer timer;
@@ -1671,9 +1671,9 @@ int main(int RAJA_UNUSED_ARG(argc), char **RAJA_UNUSED_ARG(argv[]))
                           cudaMemcpyDeviceToHost ) );
 
   // Reset data in Views to CPU data
-  L.set_data(L_data); 
-  psi.set_data(psi_data); 
-  phi.set_data(phi_data); 
+  L.set_data(L_data);
+  psi.set_data(psi_data);
+  phi.set_data(phi_data);
   checkResult(phi, L, psi, num_m, num_d, num_g, num_z);
 #endif
 
@@ -1874,11 +1874,11 @@ int main(int RAJA_UNUSED_ARG(argc), char **RAJA_UNUSED_ARG(argv[]))
     RAJA::KernelPolicy<
       statement::HipKernelAsync<
         statement::InitLocalMem<hip_shared_mem, ParamList<0,1>,
-          // Tile outer m,d loops 
+          // Tile outer m,d loops
           statement::Tile<0, tile_fixed<tile_m>, seq_exec,  // m
             statement::Tile<1, tile_fixed<tile_d>, seq_exec,  // d
 
-              // Load L for m,d tile into shmem 
+              // Load L for m,d tile into shmem
               statement::For<1, hip_thread_x_loop,  // d
                 statement::For<0, hip_thread_y_direct,   // m
                   statement::Lambda<0, Segs<0,1>, Params<0>, Offsets<0,1>>
@@ -1915,7 +1915,7 @@ int main(int RAJA_UNUSED_ARG(argc), char **RAJA_UNUSED_ARG(argv[]))
                       > // d
                     >  // m
                   >,  // z
-                  
+
                   // finish tile over directions
                   statement::HipSyncThreads,
 
@@ -1926,7 +1926,7 @@ int main(int RAJA_UNUSED_ARG(argc), char **RAJA_UNUSED_ARG(argv[]))
                     >
                   >,
                   statement::HipSyncThreads
-                
+
                 >  // Tile z
               >  // g
 
@@ -1983,7 +1983,7 @@ int main(int RAJA_UNUSED_ARG(argc), char **RAJA_UNUSED_ARG(argv[]))
     },
 
     // Lambda<3> Compute thread-local phi value
-    [=] RAJA_DEVICE (IM RAJA_UNUSED_ARG(m), ID RAJA_UNUSED_ARG(d), 
+    [=] RAJA_DEVICE (IM RAJA_UNUSED_ARG(m), ID RAJA_UNUSED_ARG(d),
                      IG RAJA_UNUSED_ARG(g), IZ RAJA_UNUSED_ARG(z),
                      shmem_L_t& sh_L, shmem_psi_t& sh_psi, double& phi_local,
                      IM tm, ID td, IG tg, IZ tz) {
@@ -2040,7 +2040,7 @@ int main(int RAJA_UNUSED_ARG(argc), char **RAJA_UNUSED_ARG(argv[]))
 //
 template <typename PHIVIEW_T, typename LVIEW_T, typename PSIVIEW_T>
 void checkResult(PHIVIEW_T& phi, LVIEW_T& L, PSIVIEW_T& psi,
-                 const int num_m, 
+                 const int num_m,
                  const int num_d,
                  const int num_g,
                  const int num_z)
diff --git a/scripts/lc-builds/blueos_nvcc_gcc.sh b/scripts/lc-builds/blueos_nvcc_gcc.sh
index 7619c360d4..60e1fadcd0 100755
--- a/scripts/lc-builds/blueos_nvcc_gcc.sh
+++ b/scripts/lc-builds/blueos_nvcc_gcc.sh
@@ -44,6 +44,7 @@ cmake \
   -C ../host-configs/lc-builds/blueos/nvcc_gcc_X.cmake \
   -DENABLE_OPENMP=On \
   -DENABLE_CUDA=On \
+  -DENABLE_BENCHMARKS=On \
   -DCUDA_TOOLKIT_ROOT_DIR=/usr/tce/packages/cuda/cuda-${COMP_NVCC_VER} \
   -DCMAKE_CUDA_COMPILER=/usr/tce/packages/cuda/cuda-${COMP_NVCC_VER}/bin/nvcc \
   -DCMAKE_CUDA_ARCHITECTURES=${COMP_ARCH} \
diff --git a/scripts/lc-builds/toss4_amdclang.sh b/scripts/lc-builds/toss4_amdclang.sh
index 9e738fc781..cb5872c0c5 100755
--- a/scripts/lc-builds/toss4_amdclang.sh
+++ b/scripts/lc-builds/toss4_amdclang.sh
@@ -73,6 +73,7 @@ cmake \
   -DENABLE_HIP=ON \
   -DENABLE_OPENMP=ON \
   -DENABLE_CUDA=OFF \
+  -DENABLE_BENCHMARKS=ON \
   -DCMAKE_INSTALL_PREFIX=../install_${BUILD_SUFFIX} \
   "$@" \
   ..

From 887d3d3b744fababba3b5e61d185e2d5d1717c9d Mon Sep 17 00:00:00 2001
From: Robert Chen <chen59@llnl.gov>
Date: Fri, 27 Sep 2024 15:55:07 -0700
Subject: [PATCH 2/8] Add strongly typed index support for TensorIndex.

---
 include/RAJA/pattern/tensor/TensorIndex.hpp | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/include/RAJA/pattern/tensor/TensorIndex.hpp b/include/RAJA/pattern/tensor/TensorIndex.hpp
index f992649876..379e13ae41 100644
--- a/include/RAJA/pattern/tensor/TensorIndex.hpp
+++ b/include/RAJA/pattern/tensor/TensorIndex.hpp
@@ -29,7 +29,7 @@ namespace expt
 {
 
 
-  template<typename IDX, typename TENSOR_TYPE, camp::idx_t DIM, IDX INDEX_VALUE, strip_index_type_t<IDX> LENGTH_VALUE>
+  template<typename IDX, typename TENSOR_TYPE, camp::idx_t DIM, strip_index_type_t<IDX> INDEX_VALUE, strip_index_type_t<IDX> LENGTH_VALUE>
   struct StaticTensorIndexInner;
 
   template<typename INNER_TYPE>
@@ -56,8 +56,8 @@ namespace expt
       RAJA_HOST_DEVICE
       static
       constexpr
-      StaticTensorIndex<StaticTensorIndexInner<IDX,TENSOR_TYPE,DIM,index_type(-1),value_type(-1)>> static_all(){
-        return StaticTensorIndex<StaticTensorIndexInner<IDX,TENSOR_TYPE,DIM,index_type(-1),value_type(-1)>>();
+      StaticTensorIndex<StaticTensorIndexInner<IDX,TENSOR_TYPE,DIM,value_type(-1),value_type(-1)>> static_all(){
+        return StaticTensorIndex<StaticTensorIndexInner<IDX,TENSOR_TYPE,DIM,value_type(-1),value_type(-1)>>();
       }
 
       RAJA_INLINE
@@ -103,7 +103,7 @@ namespace expt
       TensorIndex(TensorIndex<IDX, T, D> const &c) : m_index(*c), m_length(c.size()) {}
 
 
-      template<IDX IDX_VAL, strip_index_type_t<IDX> LEN_VAL>
+      template<strip_index_type_t<IDX> IDX_VAL, strip_index_type_t<IDX> LEN_VAL>
       RAJA_INLINE
       RAJA_HOST_DEVICE
       constexpr
@@ -156,7 +156,7 @@ namespace expt
   };
 
 
-  template<typename IDX, typename TENSOR_TYPE, camp::idx_t DIM, IDX INDEX_VALUE, strip_index_type_t<IDX> LENGTH_VALUE>
+  template<typename IDX, typename TENSOR_TYPE, camp::idx_t DIM, strip_index_type_t<IDX> INDEX_VALUE, strip_index_type_t<IDX> LENGTH_VALUE>
   struct StaticTensorIndex<StaticTensorIndexInner<IDX,TENSOR_TYPE,DIM,INDEX_VALUE,LENGTH_VALUE>> {
 
       using base_type  = TensorIndex<IDX,TENSOR_TYPE,DIM>;

From 5a8ee0650ece6218ad61b10f2e33cf1c07b9e5b8 Mon Sep 17 00:00:00 2001
From: Robert Chen <chen59@llnl.gov>
Date: Fri, 27 Sep 2024 16:18:04 -0700
Subject: [PATCH 3/8] Fix allocating too much to std::vector.

---
 benchmark/ltimes.cpp | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/benchmark/ltimes.cpp b/benchmark/ltimes.cpp
index c85826eff8..15a3e54eb8 100644
--- a/benchmark/ltimes.cpp
+++ b/benchmark/ltimes.cpp
@@ -142,7 +142,7 @@ int main(int RAJA_UNUSED_ARG(argc), char **RAJA_UNUSED_ARG(argv[]))
   const long num_z = 32 + (rand()/RAND_MAX);
 #else
   const int num_iter = 10 + (rand()/RAND_MAX);
-  const int num_z = 32*65536 + (rand()/RAND_MAX);
+  const int num_z = 32*657 + (rand()/RAND_MAX);
 
 #endif
 

From 84d9c9f34685c9680affa99db57ad09af169adce Mon Sep 17 00:00:00 2001
From: Arturo Vargas <vargas45@llnl.gov>
Date: Wed, 9 Oct 2024 11:18:00 -0700
Subject: [PATCH 4/8] optional benchmark

---
 scripts/lc-builds/blueos_nvcc_gcc.sh | 18 +++++++++++++-----
 1 file changed, 13 insertions(+), 5 deletions(-)

diff --git a/scripts/lc-builds/blueos_nvcc_gcc.sh b/scripts/lc-builds/blueos_nvcc_gcc.sh
index 60e1fadcd0..40a67205c1 100755
--- a/scripts/lc-builds/blueos_nvcc_gcc.sh
+++ b/scripts/lc-builds/blueos_nvcc_gcc.sh
@@ -9,22 +9,30 @@
 
 if [[ $# -lt 3 ]]; then
   echo
-  echo "You must pass 3 arguments to the script (in this order): "
+  echo "You must pass at least 3 arguments to the script (in this order): "
   echo "   1) compiler version number for nvcc"
   echo "   2) CUDA compute architecture (number only, not 'sm_70' for example)"
   echo "   3) compiler version number for gcc"
+  echo "   4) (optional) benchmarks On"
   echo
   echo "For example: "
-  echo "    blueos_nvcc_gcc.sh 10.2.89 70 8.3.1"
+  echo "    blueos_nvcc_gcc.sh 10.2.89 70 8.3.1 On"
   exit
 fi
 
 COMP_NVCC_VER=$1
 COMP_ARCH=$2
 COMP_GCC_VER=$3
-shift 3
 
-BUILD_SUFFIX=lc_blueos-nvcc${COMP_NVCC_VER}-${COMP_ARCH}-gcc${COMP_GCC_VER}
+if [[ $# -gt 3 ]]; then 
+  BENCHMARKS=$4
+  shift 4
+  BUILD_SUFFIX=lc_blueos-nvcc${COMP_NVCC_VER}-${COMP_ARCH}-gcc${COMP_GCC_VER}-benchmark_${BENCHMARKS}
+else
+  shift 3
+  BENCHMARKS=Off
+  BUILD_SUFFIX=lc_blueos-nvcc${COMP_NVCC_VER}-${COMP_ARCH}-gcc${COMP_GCC_VER}
+fi
 
 echo
 echo "Creating build directory build_${BUILD_SUFFIX} and generating configuration in it"
@@ -44,7 +52,7 @@ cmake \
   -C ../host-configs/lc-builds/blueos/nvcc_gcc_X.cmake \
   -DENABLE_OPENMP=On \
   -DENABLE_CUDA=On \
-  -DENABLE_BENCHMARKS=On \
+  -DENABLE_BENCHMARKS=${BENCHMARKS} \
   -DCUDA_TOOLKIT_ROOT_DIR=/usr/tce/packages/cuda/cuda-${COMP_NVCC_VER} \
   -DCMAKE_CUDA_COMPILER=/usr/tce/packages/cuda/cuda-${COMP_NVCC_VER}/bin/nvcc \
   -DCMAKE_CUDA_ARCHITECTURES=${COMP_ARCH} \

From 4d904622df5ac6b216923d5e3a0ca29452376f4e Mon Sep 17 00:00:00 2001
From: Arturo Vargas <vargas45@llnl.gov>
Date: Wed, 9 Oct 2024 14:01:00 -0700
Subject: [PATCH 5/8] turn on benchmarks

---
 scripts/lc-builds/blueos_clang.sh           |  1 +
 scripts/lc-builds/blueos_clang_omptarget.sh |  1 +
 scripts/lc-builds/blueos_clangcuda.sh       |  1 +
 scripts/lc-builds/blueos_gcc.sh             |  1 +
 scripts/lc-builds/blueos_nvcc_clang.sh      |  1 +
 scripts/lc-builds/blueos_nvcc_gcc.sh        | 18 +++++-------------
 scripts/lc-builds/blueos_nvcc_xl.sh         |  1 +
 scripts/lc-builds/blueos_pgi.sh             |  1 +
 scripts/lc-builds/blueos_xl.sh              |  1 +
 scripts/lc-builds/blueos_xl_omptarget.sh    |  1 +
 scripts/lc-builds/corona_sycl.sh            |  1 +
 scripts/lc-builds/toss4_amdclang_asan.sh    |  1 +
 scripts/lc-builds/toss4_cce_hip.sh          |  1 +
 scripts/lc-builds/toss4_clang.sh            |  1 +
 scripts/lc-builds/toss4_clang_san.sh        |  1 +
 scripts/lc-builds/toss4_gcc.sh              |  1 +
 scripts/lc-builds/toss4_hipcc.sh            |  1 +
 scripts/lc-builds/toss4_icpc-classic.sh     |  1 +
 scripts/lc-builds/toss4_icpc.sh             |  1 +
 scripts/lc-builds/toss4_icpx.sh             |  1 +
 20 files changed, 24 insertions(+), 13 deletions(-)

diff --git a/scripts/lc-builds/blueos_clang.sh b/scripts/lc-builds/blueos_clang.sh
index 7e0b6600d8..b511977e52 100755
--- a/scripts/lc-builds/blueos_clang.sh
+++ b/scripts/lc-builds/blueos_clang.sh
@@ -38,6 +38,7 @@ cmake \
   -DBLT_CXX_STD=c++14 \
   -C ../host-configs/lc-builds/blueos/clang_X.cmake \
   -DENABLE_OPENMP=On \
+  -DENABLE_BENCHMARKS=ON \
   -DCMAKE_INSTALL_PREFIX=../install_${BUILD_SUFFIX} \
   "$@" \
   ..
diff --git a/scripts/lc-builds/blueos_clang_omptarget.sh b/scripts/lc-builds/blueos_clang_omptarget.sh
index 154a1f764a..2f5ae90e87 100755
--- a/scripts/lc-builds/blueos_clang_omptarget.sh
+++ b/scripts/lc-builds/blueos_clang_omptarget.sh
@@ -42,6 +42,7 @@ cmake \
   -DRAJA_ENABLE_TARGET_OPENMP=On \
   -DBLT_OPENMP_COMPILE_FLAGS="-fopenmp;-fopenmp-targets=nvptx64-nvidia-cuda" \
   -DBLT_OPENMP_LINK_FLAGS="-fopenmp;-fopenmp-targets=nvptx64-nvidia-cuda" \
+  -DENABLE_BENCHMARKS=ON \
   -DCMAKE_INSTALL_PREFIX=../install_${BUILD_SUFFIX} \
   "$@" \
   ..
diff --git a/scripts/lc-builds/blueos_clangcuda.sh b/scripts/lc-builds/blueos_clangcuda.sh
index 16ac65b825..00e230fb43 100755
--- a/scripts/lc-builds/blueos_clangcuda.sh
+++ b/scripts/lc-builds/blueos_clangcuda.sh
@@ -50,6 +50,7 @@ cmake \
   -DBLT_CLANG_CUDA_ARCH=${CUDA_ARCH} \
   -DENABLE_CUDA=On \
   -DCUDA_ARCH=${CUDA_ARCH} \
+  -DENABLE_BENCHMARKS=ON \
   -DCMAKE_INSTALL_PREFIX=../install_${BUILD_SUFFIX} \
   "$@" \
   ..
diff --git a/scripts/lc-builds/blueos_gcc.sh b/scripts/lc-builds/blueos_gcc.sh
index 65bfb2c307..a87fd91fab 100755
--- a/scripts/lc-builds/blueos_gcc.sh
+++ b/scripts/lc-builds/blueos_gcc.sh
@@ -36,6 +36,7 @@ cmake \
   -DBLT_CXX_STD=c++14 \
   -C ../host-configs/lc-builds/blueos/gcc_X.cmake \
   -DENABLE_OPENMP=On \
+  -DENABLE_BENCHMARKS=ON \
   -DCMAKE_INSTALL_PREFIX=../install_${BUILD_SUFFIX} \
   "$@" \
   ..
diff --git a/scripts/lc-builds/blueos_nvcc_clang.sh b/scripts/lc-builds/blueos_nvcc_clang.sh
index faf0c8938a..82f70ba38e 100755
--- a/scripts/lc-builds/blueos_nvcc_clang.sh
+++ b/scripts/lc-builds/blueos_nvcc_clang.sh
@@ -48,6 +48,7 @@ cmake \
   -DCUDA_TOOLKIT_ROOT_DIR=/usr/tce/packages/cuda/cuda-${COMP_NVCC_VER} \
   -DCMAKE_CUDA_COMPILER=/usr/tce/packages/cuda/cuda-${COMP_NVCC_VER}/bin/nvcc \
   -DCMAKE_CUDA_ARCHITECTURES=${COMP_ARCH} \
+  -DENABLE_BENCHMARKS=ON \
   -DCMAKE_INSTALL_PREFIX=../install_${BUILD_SUFFIX} \
   "$@" \
   ..
diff --git a/scripts/lc-builds/blueos_nvcc_gcc.sh b/scripts/lc-builds/blueos_nvcc_gcc.sh
index 40a67205c1..1b84321f1b 100755
--- a/scripts/lc-builds/blueos_nvcc_gcc.sh
+++ b/scripts/lc-builds/blueos_nvcc_gcc.sh
@@ -9,30 +9,22 @@
 
 if [[ $# -lt 3 ]]; then
   echo
-  echo "You must pass at least 3 arguments to the script (in this order): "
+  echo "You must pass 3 arguments to the script (in this order): "
   echo "   1) compiler version number for nvcc"
   echo "   2) CUDA compute architecture (number only, not 'sm_70' for example)"
   echo "   3) compiler version number for gcc"
-  echo "   4) (optional) benchmarks On"
   echo
   echo "For example: "
-  echo "    blueos_nvcc_gcc.sh 10.2.89 70 8.3.1 On"
+  echo "    blueos_nvcc_gcc.sh 10.2.89 70 8.3.1"
   exit
 fi
 
 COMP_NVCC_VER=$1
 COMP_ARCH=$2
 COMP_GCC_VER=$3
+shift 3
 
-if [[ $# -gt 3 ]]; then 
-  BENCHMARKS=$4
-  shift 4
-  BUILD_SUFFIX=lc_blueos-nvcc${COMP_NVCC_VER}-${COMP_ARCH}-gcc${COMP_GCC_VER}-benchmark_${BENCHMARKS}
-else
-  shift 3
-  BENCHMARKS=Off
-  BUILD_SUFFIX=lc_blueos-nvcc${COMP_NVCC_VER}-${COMP_ARCH}-gcc${COMP_GCC_VER}
-fi
+BUILD_SUFFIX=lc_blueos-nvcc${COMP_NVCC_VER}-${COMP_ARCH}-gcc${COMP_GCC_VER}
 
 echo
 echo "Creating build directory build_${BUILD_SUFFIX} and generating configuration in it"
@@ -52,10 +44,10 @@ cmake \
   -C ../host-configs/lc-builds/blueos/nvcc_gcc_X.cmake \
   -DENABLE_OPENMP=On \
   -DENABLE_CUDA=On \
-  -DENABLE_BENCHMARKS=${BENCHMARKS} \
   -DCUDA_TOOLKIT_ROOT_DIR=/usr/tce/packages/cuda/cuda-${COMP_NVCC_VER} \
   -DCMAKE_CUDA_COMPILER=/usr/tce/packages/cuda/cuda-${COMP_NVCC_VER}/bin/nvcc \
   -DCMAKE_CUDA_ARCHITECTURES=${COMP_ARCH} \
+  -DENABLE_BENCHMARKS=ON \
   -DCMAKE_INSTALL_PREFIX=../install_${BUILD_SUFFIX} \
   "$@" \
   ..
diff --git a/scripts/lc-builds/blueos_nvcc_xl.sh b/scripts/lc-builds/blueos_nvcc_xl.sh
index d0507d0eb1..142b3fbd40 100755
--- a/scripts/lc-builds/blueos_nvcc_xl.sh
+++ b/scripts/lc-builds/blueos_nvcc_xl.sh
@@ -47,6 +47,7 @@ cmake \
   -DCUDA_TOOLKIT_ROOT_DIR=/usr/tce/packages/cuda/cuda-${COMP_NVCC_VER} \
   -DCMAKE_CUDA_COMPILER=/usr/tce/packages/cuda/cuda-${COMP_NVCC_VER}/bin/nvcc \
   -DCMAKE_CUDA_ARCHITECTURES=${COMP_ARCH} \
+  -DENABLE_BENCHMARKS=ON \
   -DCMAKE_INSTALL_PREFIX=../install_${BUILD_SUFFIX} \
   "$@" \
   ..
diff --git a/scripts/lc-builds/blueos_pgi.sh b/scripts/lc-builds/blueos_pgi.sh
index 16c9816d19..743f7afbb6 100755
--- a/scripts/lc-builds/blueos_pgi.sh
+++ b/scripts/lc-builds/blueos_pgi.sh
@@ -37,6 +37,7 @@ cmake \
   -DBLT_CXX_STD=c++14 \
   -C ../host-configs/lc-builds/blueos/pgi_X.cmake \
   -DENABLE_OPENMP=On \
+  -DENABLE_BENCHMARKS=ON \
   -DCMAKE_INSTALL_PREFIX=../install_${BUILD_SUFFIX} \
   "$@" \
   ..
diff --git a/scripts/lc-builds/blueos_xl.sh b/scripts/lc-builds/blueos_xl.sh
index 58a13fc065..08e693023b 100755
--- a/scripts/lc-builds/blueos_xl.sh
+++ b/scripts/lc-builds/blueos_xl.sh
@@ -36,6 +36,7 @@ cmake \
   -DBLT_CXX_STD=c++14 \
   -C ../host-configs/lc-builds/blueos/xl_X.cmake \
   -DENABLE_OPENMP=On \
+  -DENABLE_BENCHMARKS=ON \
   -DCMAKE_INSTALL_PREFIX=../install_${BUILD_SUFFIX} \
   "$@" \
   ..
diff --git a/scripts/lc-builds/blueos_xl_omptarget.sh b/scripts/lc-builds/blueos_xl_omptarget.sh
index 3fd7e22ce8..cd7d9d67c6 100755
--- a/scripts/lc-builds/blueos_xl_omptarget.sh
+++ b/scripts/lc-builds/blueos_xl_omptarget.sh
@@ -39,6 +39,7 @@ cmake \
   -DRAJA_ENABLE_TARGET_OPENMP=On \
   -DBLT_OPENMP_COMPILE_FLAGS="-qoffload;-qsmp=omp;-qalias=noansi" \
   -DBLT_OPENMP_LINK_FLAGS="-qoffload;-qsmp=omp;-qalias=noansi" \
+  -DENABLE_BENCHMARKS=ON \
   -DCMAKE_INSTALL_PREFIX=../install_${BUILD_SUFFIX} \
   "$@" \
   ..
diff --git a/scripts/lc-builds/corona_sycl.sh b/scripts/lc-builds/corona_sycl.sh
index 815928e434..cf0997df65 100755
--- a/scripts/lc-builds/corona_sycl.sh
+++ b/scripts/lc-builds/corona_sycl.sh
@@ -55,6 +55,7 @@ cmake \
   -DBLT_CXX_STD=c++17 \
   -DENABLE_TESTS=On \
   -DENABLE_EXAMPLES=On \
+  -DENABLE_BENCHMARKS=ON \
   "$@" \
   ..
 
diff --git a/scripts/lc-builds/toss4_amdclang_asan.sh b/scripts/lc-builds/toss4_amdclang_asan.sh
index b22cdce3bd..e7fc4045e5 100755
--- a/scripts/lc-builds/toss4_amdclang_asan.sh
+++ b/scripts/lc-builds/toss4_amdclang_asan.sh
@@ -83,6 +83,7 @@ cmake \
   -DENABLE_HIP=ON \
   -DENABLE_OPENMP=ON \
   -DENABLE_CUDA=OFF \
+  -DENABLE_BENCHMARKS=ON \
   -DCMAKE_INSTALL_PREFIX=../install_${BUILD_SUFFIX} \
   "$@" \
   ..
diff --git a/scripts/lc-builds/toss4_cce_hip.sh b/scripts/lc-builds/toss4_cce_hip.sh
index 80db5f37be..826fadf5fa 100755
--- a/scripts/lc-builds/toss4_cce_hip.sh
+++ b/scripts/lc-builds/toss4_cce_hip.sh
@@ -57,6 +57,7 @@ cmake \
   -DENABLE_HIP=ON \
   -DENABLE_OPENMP=ON \
   -DENABLE_CUDA=OFF \
+  -DENABLE_BENCHMARKS=ON \
   -DCMAKE_INSTALL_PREFIX=../install_${BUILD_SUFFIX} \
   "$@" \
   ..
diff --git a/scripts/lc-builds/toss4_clang.sh b/scripts/lc-builds/toss4_clang.sh
index ab1cb69e81..d031b1ae7e 100755
--- a/scripts/lc-builds/toss4_clang.sh
+++ b/scripts/lc-builds/toss4_clang.sh
@@ -36,6 +36,7 @@ cmake \
   -DBLT_CXX_STD=c++14 \
   -C ../host-configs/lc-builds/toss4/clang_X.cmake \
   -DENABLE_OPENMP=On \
+  -DENABLE_BENCHMARKS=ON \
   -DCMAKE_INSTALL_PREFIX=../install_${BUILD_SUFFIX} \
   "$@" \
   ..
diff --git a/scripts/lc-builds/toss4_clang_san.sh b/scripts/lc-builds/toss4_clang_san.sh
index e7501c09cd..7d917f4c96 100755
--- a/scripts/lc-builds/toss4_clang_san.sh
+++ b/scripts/lc-builds/toss4_clang_san.sh
@@ -45,6 +45,7 @@ cmake \
   -DBLT_CXX_STD=c++14 \
   -C ../host-configs/lc-builds/toss4/clang_X_${SAN_VER}.cmake \
   -DENABLE_OPENMP=On \
+  -DENABLE_BENCHMARKS=ON \
   -DCMAKE_INSTALL_PREFIX=../install_${BUILD_SUFFIX} \
   "$@" \
   ..
diff --git a/scripts/lc-builds/toss4_gcc.sh b/scripts/lc-builds/toss4_gcc.sh
index 07fa26cc3f..9629cdd944 100755
--- a/scripts/lc-builds/toss4_gcc.sh
+++ b/scripts/lc-builds/toss4_gcc.sh
@@ -36,6 +36,7 @@ cmake \
   -DBLT_CXX_STD=c++14 \
   -C ../host-configs/lc-builds/toss4/gcc_X.cmake \
   -DENABLE_OPENMP=On \
+  -DENABLE_BENCHMARKS=ON \
   -DCMAKE_INSTALL_PREFIX=../install_${BUILD_SUFFIX} \
   "$@" \
   ..
diff --git a/scripts/lc-builds/toss4_hipcc.sh b/scripts/lc-builds/toss4_hipcc.sh
index f7342e474c..0cec1abc19 100755
--- a/scripts/lc-builds/toss4_hipcc.sh
+++ b/scripts/lc-builds/toss4_hipcc.sh
@@ -74,6 +74,7 @@ cmake \
   -DENABLE_HIP=ON \
   -DENABLE_OPENMP=ON \
   -DENABLE_CUDA=OFF \
+  -DENABLE_BENCHMARKS=ON \
   -DCMAKE_INSTALL_PREFIX=../install_${BUILD_SUFFIX} \
   "$@" \
   ..
diff --git a/scripts/lc-builds/toss4_icpc-classic.sh b/scripts/lc-builds/toss4_icpc-classic.sh
index 447cf35ac8..73481fd794 100755
--- a/scripts/lc-builds/toss4_icpc-classic.sh
+++ b/scripts/lc-builds/toss4_icpc-classic.sh
@@ -43,6 +43,7 @@ cmake \
   -C ../host-configs/lc-builds/toss4/icpc-classic_X.cmake \
   -DRAJA_ENABLE_FORCEINLINE_RECURSIVE=Off \
   -DENABLE_OPENMP=On \
+  -DENABLE_BENCHMARKS=ON \
   -DCMAKE_INSTALL_PREFIX=../install_${BUILD_SUFFIX} \
   "$@" \
   ..
diff --git a/scripts/lc-builds/toss4_icpc.sh b/scripts/lc-builds/toss4_icpc.sh
index 93d4d62cad..921a8dad54 100755
--- a/scripts/lc-builds/toss4_icpc.sh
+++ b/scripts/lc-builds/toss4_icpc.sh
@@ -43,6 +43,7 @@ cmake \
   -C ../host-configs/lc-builds/toss4/icpc_X.cmake \
   -DRAJA_ENABLE_FORCEINLINE_RECURSIVE=Off \
   -DENABLE_OPENMP=On \
+  -DENABLE_BENCHMARKS=ON \
   -DCMAKE_INSTALL_PREFIX=../install_${BUILD_SUFFIX} \
   "$@" \
   ..
diff --git a/scripts/lc-builds/toss4_icpx.sh b/scripts/lc-builds/toss4_icpx.sh
index d6fe0d867b..2f7f768a60 100755
--- a/scripts/lc-builds/toss4_icpx.sh
+++ b/scripts/lc-builds/toss4_icpx.sh
@@ -45,6 +45,7 @@ cmake \
   -C ../host-configs/lc-builds/toss4/icpx_X.cmake \
   -DRAJA_ENABLE_FORCEINLINE_RECURSIVE=Off \
   -DENABLE_OPENMP=On \
+  -DENABLE_BENCHMARKS=ON \
   -DCMAKE_INSTALL_PREFIX=../install_${BUILD_SUFFIX} \
   "$@" \
   ..

From a8c6c3813e43fa3ae530df2d2e870b83118b421e Mon Sep 17 00:00:00 2001
From: Arturo Vargas <vargas45@llnl.gov>
Date: Wed, 9 Oct 2024 14:06:09 -0700
Subject: [PATCH 6/8] Use On instead of ON

---
 scripts/lc-builds/blueos_clang.sh           | 2 +-
 scripts/lc-builds/blueos_clang_omptarget.sh | 2 +-
 scripts/lc-builds/blueos_clangcuda.sh       | 2 +-
 scripts/lc-builds/blueos_gcc.sh             | 2 +-
 scripts/lc-builds/blueos_nvcc_clang.sh      | 2 +-
 scripts/lc-builds/blueos_nvcc_gcc.sh        | 2 +-
 scripts/lc-builds/blueos_nvcc_xl.sh         | 2 +-
 scripts/lc-builds/blueos_pgi.sh             | 2 +-
 scripts/lc-builds/blueos_xl.sh              | 2 +-
 scripts/lc-builds/blueos_xl_omptarget.sh    | 2 +-
 scripts/lc-builds/corona_sycl.sh            | 2 +-
 scripts/lc-builds/toss4_amdclang.sh         | 2 +-
 scripts/lc-builds/toss4_amdclang_asan.sh    | 2 +-
 scripts/lc-builds/toss4_cce_hip.sh          | 2 +-
 scripts/lc-builds/toss4_clang.sh            | 2 +-
 scripts/lc-builds/toss4_gcc.sh              | 2 +-
 scripts/lc-builds/toss4_hipcc.sh            | 2 +-
 scripts/lc-builds/toss4_icpc-classic.sh     | 2 +-
 scripts/lc-builds/toss4_icpc.sh             | 2 +-
 scripts/lc-builds/toss4_icpx.sh             | 2 +-
 20 files changed, 20 insertions(+), 20 deletions(-)

diff --git a/scripts/lc-builds/blueos_clang.sh b/scripts/lc-builds/blueos_clang.sh
index b511977e52..99954b3eb8 100755
--- a/scripts/lc-builds/blueos_clang.sh
+++ b/scripts/lc-builds/blueos_clang.sh
@@ -38,7 +38,7 @@ cmake \
   -DBLT_CXX_STD=c++14 \
   -C ../host-configs/lc-builds/blueos/clang_X.cmake \
   -DENABLE_OPENMP=On \
-  -DENABLE_BENCHMARKS=ON \
+  -DENABLE_BENCHMARKS=On \
   -DCMAKE_INSTALL_PREFIX=../install_${BUILD_SUFFIX} \
   "$@" \
   ..
diff --git a/scripts/lc-builds/blueos_clang_omptarget.sh b/scripts/lc-builds/blueos_clang_omptarget.sh
index 2f5ae90e87..977e0ec4a3 100755
--- a/scripts/lc-builds/blueos_clang_omptarget.sh
+++ b/scripts/lc-builds/blueos_clang_omptarget.sh
@@ -42,7 +42,7 @@ cmake \
   -DRAJA_ENABLE_TARGET_OPENMP=On \
   -DBLT_OPENMP_COMPILE_FLAGS="-fopenmp;-fopenmp-targets=nvptx64-nvidia-cuda" \
   -DBLT_OPENMP_LINK_FLAGS="-fopenmp;-fopenmp-targets=nvptx64-nvidia-cuda" \
-  -DENABLE_BENCHMARKS=ON \
+  -DENABLE_BENCHMARKS=On \
   -DCMAKE_INSTALL_PREFIX=../install_${BUILD_SUFFIX} \
   "$@" \
   ..
diff --git a/scripts/lc-builds/blueos_clangcuda.sh b/scripts/lc-builds/blueos_clangcuda.sh
index 00e230fb43..98d2619546 100755
--- a/scripts/lc-builds/blueos_clangcuda.sh
+++ b/scripts/lc-builds/blueos_clangcuda.sh
@@ -50,7 +50,7 @@ cmake \
   -DBLT_CLANG_CUDA_ARCH=${CUDA_ARCH} \
   -DENABLE_CUDA=On \
   -DCUDA_ARCH=${CUDA_ARCH} \
-  -DENABLE_BENCHMARKS=ON \
+  -DENABLE_BENCHMARKS=On \
   -DCMAKE_INSTALL_PREFIX=../install_${BUILD_SUFFIX} \
   "$@" \
   ..
diff --git a/scripts/lc-builds/blueos_gcc.sh b/scripts/lc-builds/blueos_gcc.sh
index a87fd91fab..f89ad3e6bf 100755
--- a/scripts/lc-builds/blueos_gcc.sh
+++ b/scripts/lc-builds/blueos_gcc.sh
@@ -36,7 +36,7 @@ cmake \
   -DBLT_CXX_STD=c++14 \
   -C ../host-configs/lc-builds/blueos/gcc_X.cmake \
   -DENABLE_OPENMP=On \
-  -DENABLE_BENCHMARKS=ON \
+  -DENABLE_BENCHMARKS=On \
   -DCMAKE_INSTALL_PREFIX=../install_${BUILD_SUFFIX} \
   "$@" \
   ..
diff --git a/scripts/lc-builds/blueos_nvcc_clang.sh b/scripts/lc-builds/blueos_nvcc_clang.sh
index 82f70ba38e..ad846fa101 100755
--- a/scripts/lc-builds/blueos_nvcc_clang.sh
+++ b/scripts/lc-builds/blueos_nvcc_clang.sh
@@ -48,7 +48,7 @@ cmake \
   -DCUDA_TOOLKIT_ROOT_DIR=/usr/tce/packages/cuda/cuda-${COMP_NVCC_VER} \
   -DCMAKE_CUDA_COMPILER=/usr/tce/packages/cuda/cuda-${COMP_NVCC_VER}/bin/nvcc \
   -DCMAKE_CUDA_ARCHITECTURES=${COMP_ARCH} \
-  -DENABLE_BENCHMARKS=ON \
+  -DENABLE_BENCHMARKS=On \
   -DCMAKE_INSTALL_PREFIX=../install_${BUILD_SUFFIX} \
   "$@" \
   ..
diff --git a/scripts/lc-builds/blueos_nvcc_gcc.sh b/scripts/lc-builds/blueos_nvcc_gcc.sh
index 1b84321f1b..141936aad8 100755
--- a/scripts/lc-builds/blueos_nvcc_gcc.sh
+++ b/scripts/lc-builds/blueos_nvcc_gcc.sh
@@ -47,7 +47,7 @@ cmake \
   -DCUDA_TOOLKIT_ROOT_DIR=/usr/tce/packages/cuda/cuda-${COMP_NVCC_VER} \
   -DCMAKE_CUDA_COMPILER=/usr/tce/packages/cuda/cuda-${COMP_NVCC_VER}/bin/nvcc \
   -DCMAKE_CUDA_ARCHITECTURES=${COMP_ARCH} \
-  -DENABLE_BENCHMARKS=ON \
+  -DENABLE_BENCHMARKS=On \
   -DCMAKE_INSTALL_PREFIX=../install_${BUILD_SUFFIX} \
   "$@" \
   ..
diff --git a/scripts/lc-builds/blueos_nvcc_xl.sh b/scripts/lc-builds/blueos_nvcc_xl.sh
index 142b3fbd40..ffeea779ca 100755
--- a/scripts/lc-builds/blueos_nvcc_xl.sh
+++ b/scripts/lc-builds/blueos_nvcc_xl.sh
@@ -47,7 +47,7 @@ cmake \
   -DCUDA_TOOLKIT_ROOT_DIR=/usr/tce/packages/cuda/cuda-${COMP_NVCC_VER} \
   -DCMAKE_CUDA_COMPILER=/usr/tce/packages/cuda/cuda-${COMP_NVCC_VER}/bin/nvcc \
   -DCMAKE_CUDA_ARCHITECTURES=${COMP_ARCH} \
-  -DENABLE_BENCHMARKS=ON \
+  -DENABLE_BENCHMARKS=On \
   -DCMAKE_INSTALL_PREFIX=../install_${BUILD_SUFFIX} \
   "$@" \
   ..
diff --git a/scripts/lc-builds/blueos_pgi.sh b/scripts/lc-builds/blueos_pgi.sh
index 743f7afbb6..d8a6b904ec 100755
--- a/scripts/lc-builds/blueos_pgi.sh
+++ b/scripts/lc-builds/blueos_pgi.sh
@@ -37,7 +37,7 @@ cmake \
   -DBLT_CXX_STD=c++14 \
   -C ../host-configs/lc-builds/blueos/pgi_X.cmake \
   -DENABLE_OPENMP=On \
-  -DENABLE_BENCHMARKS=ON \
+  -DENABLE_BENCHMARKS=On \
   -DCMAKE_INSTALL_PREFIX=../install_${BUILD_SUFFIX} \
   "$@" \
   ..
diff --git a/scripts/lc-builds/blueos_xl.sh b/scripts/lc-builds/blueos_xl.sh
index 08e693023b..a39a0f4c93 100755
--- a/scripts/lc-builds/blueos_xl.sh
+++ b/scripts/lc-builds/blueos_xl.sh
@@ -36,7 +36,7 @@ cmake \
   -DBLT_CXX_STD=c++14 \
   -C ../host-configs/lc-builds/blueos/xl_X.cmake \
   -DENABLE_OPENMP=On \
-  -DENABLE_BENCHMARKS=ON \
+  -DENABLE_BENCHMARKS=On \
   -DCMAKE_INSTALL_PREFIX=../install_${BUILD_SUFFIX} \
   "$@" \
   ..
diff --git a/scripts/lc-builds/blueos_xl_omptarget.sh b/scripts/lc-builds/blueos_xl_omptarget.sh
index cd7d9d67c6..d331044a99 100755
--- a/scripts/lc-builds/blueos_xl_omptarget.sh
+++ b/scripts/lc-builds/blueos_xl_omptarget.sh
@@ -39,7 +39,7 @@ cmake \
   -DRAJA_ENABLE_TARGET_OPENMP=On \
   -DBLT_OPENMP_COMPILE_FLAGS="-qoffload;-qsmp=omp;-qalias=noansi" \
   -DBLT_OPENMP_LINK_FLAGS="-qoffload;-qsmp=omp;-qalias=noansi" \
-  -DENABLE_BENCHMARKS=ON \
+  -DENABLE_BENCHMARKS=On \
   -DCMAKE_INSTALL_PREFIX=../install_${BUILD_SUFFIX} \
   "$@" \
   ..
diff --git a/scripts/lc-builds/corona_sycl.sh b/scripts/lc-builds/corona_sycl.sh
index cf0997df65..4b636e4da0 100755
--- a/scripts/lc-builds/corona_sycl.sh
+++ b/scripts/lc-builds/corona_sycl.sh
@@ -55,7 +55,7 @@ cmake \
   -DBLT_CXX_STD=c++17 \
   -DENABLE_TESTS=On \
   -DENABLE_EXAMPLES=On \
-  -DENABLE_BENCHMARKS=ON \
+  -DENABLE_BENCHMARKS=On \
   "$@" \
   ..
 
diff --git a/scripts/lc-builds/toss4_amdclang.sh b/scripts/lc-builds/toss4_amdclang.sh
index cb5872c0c5..ec7910e148 100755
--- a/scripts/lc-builds/toss4_amdclang.sh
+++ b/scripts/lc-builds/toss4_amdclang.sh
@@ -73,7 +73,7 @@ cmake \
   -DENABLE_HIP=ON \
   -DENABLE_OPENMP=ON \
   -DENABLE_CUDA=OFF \
-  -DENABLE_BENCHMARKS=ON \
+  -DENABLE_BENCHMARKS=On \
   -DCMAKE_INSTALL_PREFIX=../install_${BUILD_SUFFIX} \
   "$@" \
   ..
diff --git a/scripts/lc-builds/toss4_amdclang_asan.sh b/scripts/lc-builds/toss4_amdclang_asan.sh
index e7fc4045e5..dd9526a1d2 100755
--- a/scripts/lc-builds/toss4_amdclang_asan.sh
+++ b/scripts/lc-builds/toss4_amdclang_asan.sh
@@ -83,7 +83,7 @@ cmake \
   -DENABLE_HIP=ON \
   -DENABLE_OPENMP=ON \
   -DENABLE_CUDA=OFF \
-  -DENABLE_BENCHMARKS=ON \
+  -DENABLE_BENCHMARKS=On \
   -DCMAKE_INSTALL_PREFIX=../install_${BUILD_SUFFIX} \
   "$@" \
   ..
diff --git a/scripts/lc-builds/toss4_cce_hip.sh b/scripts/lc-builds/toss4_cce_hip.sh
index 826fadf5fa..9d05fddca8 100755
--- a/scripts/lc-builds/toss4_cce_hip.sh
+++ b/scripts/lc-builds/toss4_cce_hip.sh
@@ -57,7 +57,7 @@ cmake \
   -DENABLE_HIP=ON \
   -DENABLE_OPENMP=ON \
   -DENABLE_CUDA=OFF \
-  -DENABLE_BENCHMARKS=ON \
+  -DENABLE_BENCHMARKS=On \
   -DCMAKE_INSTALL_PREFIX=../install_${BUILD_SUFFIX} \
   "$@" \
   ..
diff --git a/scripts/lc-builds/toss4_clang.sh b/scripts/lc-builds/toss4_clang.sh
index d031b1ae7e..a1bf63ff89 100755
--- a/scripts/lc-builds/toss4_clang.sh
+++ b/scripts/lc-builds/toss4_clang.sh
@@ -36,7 +36,7 @@ cmake \
   -DBLT_CXX_STD=c++14 \
   -C ../host-configs/lc-builds/toss4/clang_X.cmake \
   -DENABLE_OPENMP=On \
-  -DENABLE_BENCHMARKS=ON \
+  -DENABLE_BENCHMARKS=On \
   -DCMAKE_INSTALL_PREFIX=../install_${BUILD_SUFFIX} \
   "$@" \
   ..
diff --git a/scripts/lc-builds/toss4_gcc.sh b/scripts/lc-builds/toss4_gcc.sh
index 9629cdd944..c73c5ba6e9 100755
--- a/scripts/lc-builds/toss4_gcc.sh
+++ b/scripts/lc-builds/toss4_gcc.sh
@@ -36,7 +36,7 @@ cmake \
   -DBLT_CXX_STD=c++14 \
   -C ../host-configs/lc-builds/toss4/gcc_X.cmake \
   -DENABLE_OPENMP=On \
-  -DENABLE_BENCHMARKS=ON \
+  -DENABLE_BENCHMARKS=On \
   -DCMAKE_INSTALL_PREFIX=../install_${BUILD_SUFFIX} \
   "$@" \
   ..
diff --git a/scripts/lc-builds/toss4_hipcc.sh b/scripts/lc-builds/toss4_hipcc.sh
index 0cec1abc19..5743692ba6 100755
--- a/scripts/lc-builds/toss4_hipcc.sh
+++ b/scripts/lc-builds/toss4_hipcc.sh
@@ -74,7 +74,7 @@ cmake \
   -DENABLE_HIP=ON \
   -DENABLE_OPENMP=ON \
   -DENABLE_CUDA=OFF \
-  -DENABLE_BENCHMARKS=ON \
+  -DENABLE_BENCHMARKS=On \
   -DCMAKE_INSTALL_PREFIX=../install_${BUILD_SUFFIX} \
   "$@" \
   ..
diff --git a/scripts/lc-builds/toss4_icpc-classic.sh b/scripts/lc-builds/toss4_icpc-classic.sh
index 73481fd794..a892d08ecf 100755
--- a/scripts/lc-builds/toss4_icpc-classic.sh
+++ b/scripts/lc-builds/toss4_icpc-classic.sh
@@ -43,7 +43,7 @@ cmake \
   -C ../host-configs/lc-builds/toss4/icpc-classic_X.cmake \
   -DRAJA_ENABLE_FORCEINLINE_RECURSIVE=Off \
   -DENABLE_OPENMP=On \
-  -DENABLE_BENCHMARKS=ON \
+  -DENABLE_BENCHMARKS=On \
   -DCMAKE_INSTALL_PREFIX=../install_${BUILD_SUFFIX} \
   "$@" \
   ..
diff --git a/scripts/lc-builds/toss4_icpc.sh b/scripts/lc-builds/toss4_icpc.sh
index 921a8dad54..82dab8e8df 100755
--- a/scripts/lc-builds/toss4_icpc.sh
+++ b/scripts/lc-builds/toss4_icpc.sh
@@ -43,7 +43,7 @@ cmake \
   -C ../host-configs/lc-builds/toss4/icpc_X.cmake \
   -DRAJA_ENABLE_FORCEINLINE_RECURSIVE=Off \
   -DENABLE_OPENMP=On \
-  -DENABLE_BENCHMARKS=ON \
+  -DENABLE_BENCHMARKS=On \
   -DCMAKE_INSTALL_PREFIX=../install_${BUILD_SUFFIX} \
   "$@" \
   ..
diff --git a/scripts/lc-builds/toss4_icpx.sh b/scripts/lc-builds/toss4_icpx.sh
index 2f7f768a60..88cc43d824 100755
--- a/scripts/lc-builds/toss4_icpx.sh
+++ b/scripts/lc-builds/toss4_icpx.sh
@@ -45,7 +45,7 @@ cmake \
   -C ../host-configs/lc-builds/toss4/icpx_X.cmake \
   -DRAJA_ENABLE_FORCEINLINE_RECURSIVE=Off \
   -DENABLE_OPENMP=On \
-  -DENABLE_BENCHMARKS=ON \
+  -DENABLE_BENCHMARKS=On \
   -DCMAKE_INSTALL_PREFIX=../install_${BUILD_SUFFIX} \
   "$@" \
   ..

From 907d4321a7bb66b173c9e27beab64fc7d6bd52d6 Mon Sep 17 00:00:00 2001
From: Robert Chen <chen59@llnl.gov>
Date: Wed, 9 Oct 2024 14:45:46 -0700
Subject: [PATCH 7/8] Use value_type where it should be used.

---
 include/RAJA/pattern/tensor/TensorIndex.hpp | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/include/RAJA/pattern/tensor/TensorIndex.hpp b/include/RAJA/pattern/tensor/TensorIndex.hpp
index 379e13ae41..8f152d92ce 100644
--- a/include/RAJA/pattern/tensor/TensorIndex.hpp
+++ b/include/RAJA/pattern/tensor/TensorIndex.hpp
@@ -103,7 +103,7 @@ namespace expt
       TensorIndex(TensorIndex<IDX, T, D> const &c) : m_index(*c), m_length(c.size()) {}
 
 
-      template<strip_index_type_t<IDX> IDX_VAL, strip_index_type_t<IDX> LEN_VAL>
+      template<value_type IDX_VAL, value_type LEN_VAL>
       RAJA_INLINE
       RAJA_HOST_DEVICE
       constexpr

From 6385af16da3e35485927208d721905791d925775 Mon Sep 17 00:00:00 2001
From: Robert Chen <chen59@llnl.gov>
Date: Wed, 9 Oct 2024 14:59:41 -0700
Subject: [PATCH 8/8] Turn on ltimes benchmark build.

---
 benchmark/CMakeLists.txt | 7 +++----
 1 file changed, 3 insertions(+), 4 deletions(-)

diff --git a/benchmark/CMakeLists.txt b/benchmark/CMakeLists.txt
index 143d6b5b98..c52e040a76 100644
--- a/benchmark/CMakeLists.txt
+++ b/benchmark/CMakeLists.txt
@@ -17,10 +17,9 @@ raja_add_benchmark(
   SOURCES benchmark-atomic.cpp)
 endif()
 
-#TODO: Fix build issues
-#raja_add_benchmark(
-#  NAME ltimes
-#  SOURCES ltimes.cpp)
+raja_add_benchmark(
+  NAME ltimes
+  SOURCES ltimes.cpp)
 
 raja_add_benchmark(
   NAME raja_view_blur