Add missing calls to increase CUDA stack limit

This fixes crashes on my system when running with cms2018.gdml in Example11 and Example14.
apt-sim · Oct 17, 2023 · 8d1ab09 · 8d1ab09
1 parent 0e51a1e
commit 8d1ab09
Show file tree

Hide file tree

Showing 2 changed files with 2 additions and 0 deletions.
diff --git a/examples/Example11/example11.cu b/examples/Example11/example11.cu
@@ -161,6 +161,7 @@ __global__ void FinishIteration(AllParticleQueues all, const GlobalScoring *scor
 void example11(const vecgeom::cxx::VPlacedVolume *world, int numParticles, double energy)
 {
   NVTXTracer tracer("Geometry");
+  COPCORE_CUDA_CHECK(vecgeom::cxx::CudaDeviceSetStackLimit(8192));
   auto &cudaManager = vecgeom::cxx::CudaManager::Instance();
   cudaManager.LoadGeometry(world);
   cudaManager.Synchronize();

diff --git a/examples/Example14/AdeptIntegration.cu b/examples/Example14/AdeptIntegration.cu
@@ -212,6 +212,7 @@ __global__ void ClearAllQueues(AllParticleQueues all)
 
 bool AdeptIntegration::InitializeGeometry(const vecgeom::cxx::VPlacedVolume *world)
 {
+  COPCORE_CUDA_CHECK(vecgeom::cxx::CudaDeviceSetStackLimit(8192));
   // Upload geometry to GPU.
   auto &cudaManager = vecgeom::cxx::CudaManager::Instance();
   cudaManager.LoadGeometry(world);