use multiple threads in reco example (#53)

* use multiple threads in reco example

* fix typos

Author: m-fila

examples/alien_reco.cpp

@@ -124,12 +124,12 @@ subtool::Task<DeviceBuffer<int>> seeding(DeviceBuffer<int> clusters,
 
     log(self) << "Found " << nSeeds << " seeds" << std::endl;
 
-    // Allocate clusters of appropiate size on device
+    // Allocate seeds of appropiate size on device
     int* d_seeds = nullptr;
     ERROR_CHECK_CUDA(cudaMallocAsync(reinterpret_cast<void**>(&d_seeds),
                                      nSeeds * sizeof(int), stream));
 
-    // Write some dummy data to the clusters buffer to simulate work
+    // Write some dummy data to the seeds buffer to simulate work
     ERROR_CHECK_CUDA(cudaMemsetAsync(d_seeds, 0, nSeeds * sizeof(int), stream));
     ERROR_CHECK_CUDA(
         cudaMemsetAsync(d_seeds, 1, nSeeds / 2 * sizeof(int), stream));
@@ -182,7 +182,7 @@ int main() {
 
     log() << "main Starting" << std::endl;
 
-    tbb::task_arena task_arena{2};
+    tbb::task_arena task_arena{2, 0};
 
     auto scheduler = [&task_arena](std::coroutine_handle<> handle) {
         task_arena.enqueue([handle]() { handle.resume(); });

examples/capy_reco.cpp

@@ -101,12 +101,12 @@ boost::capy::task<DeviceBuffer<int>> seeding(DeviceBuffer<int> clusters,
 
     log(self) << "Found " << nSeeds << " seeds" << std::endl;
 
-    // Allocate clusters of appropiate size on device
+    // Allocate seeds of appropiate size on device
     int* d_seeds = nullptr;
     ERROR_CHECK_CUDA(cudaMallocAsync(reinterpret_cast<void**>(&d_seeds),
                                      nSeeds * sizeof(int), stream));
 
-    // Write some dummy data to the clusters buffer to simulate work
+    // Write some dummy data to the seeds buffer to simulate work
     ERROR_CHECK_CUDA(cudaMemsetAsync(d_seeds, 0, nSeeds * sizeof(int), stream));
     ERROR_CHECK_CUDA(
         cudaMemsetAsync(d_seeds, 1, nSeeds / 2 * sizeof(int), stream));
@@ -159,7 +159,7 @@ int main() {
 
     log() << "main Starting" << std::endl;
 
-    auto task_arena = tbb::task_arena{2};
+    auto task_arena = tbb::task_arena{2, 0};
     auto context = TaskArenaContext(task_arena);
     auto executor = TaskArenaExecutor(context);
 

examples/exec_reco.cpp

@@ -100,12 +100,12 @@ exec::task<DeviceBuffer<int>> seeding(DeviceBuffer<int> clusters,
 
     log(self) << "Found " << nSeeds << " seeds" << std::endl;
 
-    // Allocate clusters of appropiate size on device
+    // Allocate seeds of appropiate size on device
     int* d_seeds = nullptr;
     ERROR_CHECK_CUDA(cudaMallocAsync(reinterpret_cast<void**>(&d_seeds),
                                      nSeeds * sizeof(int), stream));
 
-    // Write some dummy data to the clusters buffer to simulate work
+    // Write some dummy data to the seeds buffer to simulate work
     ERROR_CHECK_CUDA(cudaMemsetAsync(d_seeds, 0, nSeeds * sizeof(int), stream));
     ERROR_CHECK_CUDA(
         cudaMemsetAsync(d_seeds, 1, nSeeds / 2 * sizeof(int), stream));
@@ -158,7 +158,7 @@ int main() {
 
     log() << "main Starting" << std::endl;
 
-    tbb::task_arena task_arena{2};
+    tbb::task_arena task_arena{2, 0};
     execution::scheduler auto scheduler = get_scheduler(task_arena, false);
 
     {