Refactoring and new features

## New features and APIs 
- Any grid operation (halo update, and computation) is now wrapped into a Container 
- New Naming convention for grids: Grid, Span, Field, Idx, Partition 
- Introduction of partition and span tables 
- 1D partitioner used by bGrid and eGrid 
- Grid can now generate both host and device Containers 
- Vec3D objects are tagged as constant expressions 
- New set of unit tests (a full refactoring of the unit tests is in process) 

## Refactoring 
- new dGrid implementation 
- new eGrid implementation 
- new dGrid implementation to support multi-GPU (with tray support at compile time)
- removed deprecated classes 

## Temporary dropped features 
This is a list of features and tests that have been temporally been dropped. The features will be reintroduced with future PRs.
- Reduction operations
- Sub grids 
- Staggered grids
- mGrid 
- Tutorials

## Out of sync features
- The website tutorial documentation still refers to the old API and should be updated

.clang-tidy

@@ -0,0 +1,148 @@
+# Generated from CLion Inspection settings
+---
+Checks: '-*,
+bugprone-argument-comment,
+bugprone-assert-side-effect,
+bugprone-bad-signal-to-kill-thread,
+bugprone-branch-clone,
+bugprone-copy-constructor-init,
+bugprone-dangling-handle,
+bugprone-dynamic-static-initializers,
+bugprone-fold-init-type,
+bugprone-forward-declaration-namespace,
+bugprone-forwarding-reference-overload,
+bugprone-inaccurate-erase,
+bugprone-incorrect-roundings,
+bugprone-integer-division,
+bugprone-lambda-function-name,
+bugprone-macro-parentheses,
+bugprone-macro-repeated-side-effects,
+bugprone-misplaced-operator-in-strlen-in-alloc,
+bugprone-misplaced-pointer-arithmetic-in-alloc,
+bugprone-misplaced-widening-cast,
+bugprone-move-forwarding-reference,
+bugprone-multiple-statement-macro,
+bugprone-no-escape,
+bugprone-not-null-terminated-result,
+bugprone-parent-virtual-call,
+bugprone-posix-return,
+bugprone-reserved-identifier,
+bugprone-sizeof-container,
+bugprone-sizeof-expression,
+bugprone-spuriously-wake-up-functions,
+bugprone-string-constructor,
+bugprone-string-integer-assignment,
+bugprone-string-literal-with-embedded-nul,
+bugprone-suspicious-enum-usage,
+bugprone-suspicious-include,
+bugprone-suspicious-memset-usage,
+bugprone-suspicious-missing-comma,
+bugprone-suspicious-semicolon,
+bugprone-suspicious-string-compare,
+bugprone-suspicious-memory-comparison,
+bugprone-suspicious-realloc-usage,
+bugprone-swapped-arguments,
+bugprone-terminating-continue,
+bugprone-throw-keyword-missing,
+bugprone-too-small-loop-variable,
+bugprone-undefined-memory-manipulation,
+bugprone-undelegated-constructor,
+bugprone-unhandled-self-assignment,
+bugprone-unused-raii,
+bugprone-unused-return-value,
+bugprone-use-after-move,
+bugprone-virtual-near-miss,
+cert-dcl21-cpp,
+cert-dcl58-cpp,
+cert-err34-c,
+cert-err52-cpp,
+cert-err60-cpp,
+cert-flp30-c,
+cert-msc50-cpp,
+cert-msc51-cpp,
+cert-str34-c,
+cppcoreguidelines-interfaces-global-init,
+cppcoreguidelines-narrowing-conversions,
+cppcoreguidelines-pro-type-member-init,
+cppcoreguidelines-pro-type-static-cast-downcast,
+cppcoreguidelines-slicing,
+google-default-arguments,
+google-explicit-constructor,
+google-runtime-operator,
+hicpp-exception-baseclass,
+hicpp-multiway-paths-covered,
+misc-misplaced-const,
+misc-new-delete-overloads,
+misc-no-recursion,
+misc-non-copyable-objects,
+misc-throw-by-value-catch-by-reference,
+misc-unconventional-assign-operator,
+misc-uniqueptr-reset-release,
+modernize-avoid-bind,
+modernize-concat-nested-namespaces,
+modernize-deprecated-headers,
+modernize-deprecated-ios-base-aliases,
+modernize-loop-convert,
+modernize-make-shared,
+modernize-make-unique,
+modernize-pass-by-value,
+modernize-raw-string-literal,
+modernize-redundant-void-arg,
+modernize-replace-auto-ptr,
+modernize-replace-disallow-copy-and-assign-macro,
+modernize-replace-random-shuffle,
+modernize-return-braced-init-list,
+modernize-shrink-to-fit,
+modernize-unary-static-assert,
+modernize-use-auto,
+modernize-use-bool-literals,
+modernize-use-emplace,
+modernize-use-equals-default,
+modernize-use-equals-delete,
+modernize-use-nodiscard,
+modernize-use-noexcept,
+modernize-use-nullptr,
+modernize-use-override,
+modernize-use-transparent-functors,
+modernize-use-uncaught-exceptions,
+mpi-buffer-deref,
+mpi-type-mismatch,
+openmp-use-default-none,
+performance-faster-string-find,
+performance-for-range-copy,
+performance-implicit-conversion-in-loop,
+performance-inefficient-algorithm,
+performance-inefficient-string-concatenation,
+performance-inefficient-vector-operation,
+performance-move-const-arg,
+performance-move-constructor-init,
+performance-no-automatic-move,
+performance-noexcept-move-constructor,
+performance-trivially-destructible,
+performance-type-promotion-in-math-fn,
+performance-unnecessary-copy-initialization,
+performance-unnecessary-value-param,
+portability-simd-intrinsics,
+readability-avoid-const-params-in-decls,
+readability-const-return-type,
+readability-container-size-empty,
+readability-convert-member-functions-to-static,
+readability-delete-null-pointer,
+readability-deleted-default,
+readability-inconsistent-declaration-parameter-name,
+readability-make-member-function-const,
+readability-misleading-indentation,
+readability-misplaced-array-index,
+readability-non-const-parameter,
+readability-redundant-control-flow,
+readability-redundant-declaration,
+readability-redundant-function-ptr-dereference,
+readability-redundant-smartptr-get,
+readability-redundant-string-cstr,
+readability-redundant-string-init,
+readability-simplify-subscript-expr,
+readability-static-accessed-through-instance,
+readability-static-definition-in-anonymous-namespace,
+readability-string-compare,
+readability-uniqueptr-delete-release,
+readability-use-anyofallof'

CMakeLists.txt

@@ -78,9 +78,9 @@ add_subdirectory("libNeonSys")
 add_subdirectory("libNeonSet")
 add_subdirectory("libNeonDomain")
 add_subdirectory("libNeonSkeleton")
-add_subdirectory("libNeonSolver")
-add_subdirectory("tutorials")
-add_subdirectory("apps")
+#add_subdirectory("libNeonSolver")
+#add_subdirectory("tutorials")
+#add_subdirectory("apps")
 add_subdirectory("benchmarks")
 
 
@@ -111,3 +111,9 @@ message("||  CUDA ARCHS             : ${CMAKE_CUDA_ARCHITECTURES}")
 #message("||  NeonCXXFlags           : ${NeonCXXFlags}")
 #message("||  NeonCUDAFlags          : ${NeonCUDAFlags}")
 message("\\===================================================")
+# Usefull to debug cmake
+# https://jeremimucha.com/2021/01/cmake-fundamentals-part3/
+# get_target_property(interfaceLibs NeonDeveloperLib INTERFACE_COMPILE_OPTIONS)
+# set(info ${interfaceLibs}[CXX_COMPILER_ID])
+# message(STATUS "NeonDeveloperLib INTERFACE_LINK_LIBRARIES: ${info}")
+

apps/fractal/fractal.cu

@@ -70,7 +70,7 @@ int main(int argc, char** argv)
 
         Neon::Backend backend(gpu_ids, runtime);
 
-        using Grid = Neon::domain::dGrid;
+        using Grid = Neon::dGrid;
         Grid grid(
             backend, dim,
             [](const Neon::index_3d& idx) -> bool { return true; },
@@ -89,7 +89,7 @@ int main(int argc, char** argv)
         for (time = 0; time < 1000; ++time) {
             skeleton.run();
 
-            pixels.updateIO(0);
+            pixels.updateHostData(0);
             //draw_pixels(time, pixels);
         }
     }

apps/gameOfLife/gameOfLife.cu

@@ -17,7 +17,7 @@ inline void exportVTI(FieldT& voxel_1, FieldT& voxel_2, int frame_id)
     auto io = [&](int f, FieldT& voxel) {
         printf("\n Exporting Frame =%d", f);
         int precision = 4;
-        voxel.updateIO(0);
+        voxel.updateHostData(0);
         std::ostringstream oss;
         oss << std::setw(precision) << std::setfill('0') << f;
         std::string fname = "gameOfLife_" + oss.str();
@@ -117,7 +117,7 @@ int main(int argc, char** argv)
         const size_t         num_frames = 500;
 
         using T = float;
-        using Grid = Neon::domain::dGrid;
+        using Grid = Neon::dGrid;
         Grid grid(
             backend, grid_dim,
             [](const Neon::index_3d& idx) -> bool { return true; },
@@ -139,8 +139,8 @@ int main(int argc, char** argv)
             [](const Neon::index_3d&, const int&, T& val) { val = 0; });
 
 
-        voxel_1.updateCompute(0);
-        voxel_2.updateCompute(0);
+        voxel_1.updateDeviceData(0);
+        voxel_2.updateDeviceData(0);
 
 
         std::vector<Neon::set::Container> containers;

apps/lbm/lbm.cu

@@ -624,14 +624,14 @@ inline void setup(const FlowType                  flow_type,
     velocity_1.forEachActiveCell([](const Neon::index_3d&, const int& c, T& val) { val = 0; });
     velocity_2.forEachActiveCell([](const Neon::index_3d&, const int& c, T& val) { val = 0; });
 
-    lattice_1.updateCompute(0);
-    lattice_2.updateCompute(0);
-    rho_1.updateCompute(0);
-    rho_2.updateCompute(0);
-    boundary_mask.updateCompute(0);
-    center_mask.updateCompute(0);
-    velocity_1.updateCompute(0);
-    velocity_2.updateCompute(0);
+    lattice_1.updateDeviceData(0);
+    lattice_2.updateDeviceData(0);
+    rho_1.updateDeviceData(0);
+    rho_2.updateDeviceData(0);
+    boundary_mask.updateDeviceData(0);
+    center_mask.updateDeviceData(0);
+    velocity_1.updateDeviceData(0);
+    velocity_2.updateDeviceData(0);
 }
 
 
@@ -680,7 +680,7 @@ inline void run(const int                       num_frames,
 
         if (f % t == 0) {
             backend.syncAll();
-            velocity_1.updateIO(0);
+            velocity_1.updateHostData(0);
             exportVTI(save_id, velocity_1);
             printf("\n frame  %d exported", f);
             save_id++;
@@ -722,7 +722,7 @@ int main(int argc, char** argv)
         const Neon::index_3d grid_dim(dim_x, dim_y, dim_z);
         const size_t         num_frames = (DIM == 2) ? 60000 : 2000;
 
-        using Grid = Neon::domain::dGrid;
+        using Grid = Neon::dGrid;
         using dataT = float;
 
         Grid grid(

apps/poisson/poisson.cu

@@ -106,9 +106,9 @@ void testPoisson(const Neon::Backend&             backend,
 
 
     // Move data to GPU
-    u.updateCompute(0);
-    rhs.updateCompute(0);
-    bd.updateCompute(0);
+    u.updateDeviceData(0);
+    rhs.updateDeviceData(0);
+    bd.updateDeviceData(0);
 
 
     // Laplacian matvec operation
@@ -139,7 +139,7 @@ void testPoisson(const Neon::Backend&             backend,
 
     // Plotting
     printf("Updating and exporting to VTI...\n");
-    u.updateIO(0);
+    u.updateHostData(0);
     u.ioToVtk("poisson", "u");
 }
 
@@ -161,7 +161,7 @@ int main(int argc, char** agrv)
         size_t         max_iterations = 1000;
         double         tolerance = 1e-10;
 
-        testPoisson<Neon::domain::dGrid, double, 1>(backend, domain_size, bdZMin, bdZMax, max_iterations, tolerance);
+        testPoisson<Neon::dGrid, double, 1>(backend, domain_size, bdZMin, bdZMax, max_iterations, tolerance);
     }
 
     return 0;

benchmarks/CMakeLists.txt

@@ -1,3 +1,4 @@
 cmake_minimum_required(VERSION 3.19 FATAL_ERROR)
 
 add_subdirectory("lbm-lid-driven-cavity-flow")
+# add_subdirectory("lbm-flow-over-sphere")

benchmarks/lbm-flow-over-sphere/CMakeLists.txt

@@ -0,0 +1,20 @@
+cmake_minimum_required(VERSION 3.19 FATAL_ERROR)
+
+SET(APP "lbm-flow-over-sphere")
+
+file(GLOB_RECURSE SrcFiles src/*.*)
+
+add_executable(${APP} ${SrcFiles})
+
+target_link_libraries(${APP}
+		PUBLIC libNeonDomain
+		PUBLIC libNeonSkeleton)
+
+set_target_properties(${APP} PROPERTIES
+		CUDA_SEPARABLE_COMPILATION ON
+		CUDA_RESOLVE_DEVICE_SYMBOLS ON)
+
+target_compile_options(${APP} INTERFACE
+		$<$<COMPILE_LANGUAGE:CXX>:${NeonCXXFlags}>
+		$<$<COMPILE_LANGUAGE:CUDA>:${NeonCUDAFlags}>
+		)

benchmarks/lbm-flow-over-sphere/run.sh

@@ -0,0 +1,28 @@
+set -x
+
+DOMAIN_SIZE_LIST="128 192 256 320 384 448 512"
+GRID="dGrid"
+STORAGE_FP_LIST="double float"
+COMPUTE_FP_LIST="double float"
+OCC="nOCC"
+
+for DOMAIN_SIZE in ${DOMAIN_SIZE_LIST}; do
+  for STORAGE_FP in ${STORAGE_FP_LIST}; do
+    for COMPUTE_FP in ${COMPUTE_FP_LIST}; do
+
+      if [ "${STORAGE_FP}_${COMPUTE_FP}" = "double_float" ]; then
+        continue
+      fi
+
+      echo ./lbm-flow-over-cylinder \
+        --deviceType gpu --deviceIds 0 \
+        --grid "${GRID}" \
+        --domain-size "${DOMAIN_SIZE}" \
+        --warmup-iter 10 --max-iter 100 --repetitions 5 \
+        --report-filename "lbm-flow-over-cylinder_${DOMAIN_SIZE}_${GRID}_STORAGE_${STORAGE_FP}_COMPUTE_${COMPUTE_FP}" \
+        --computeFP "${COMPUTE_FP}" \
+        --storageFP "${STORAGE_FP}" \
+        --${OCC} --benchmark
+    done
+  done
+done