diff --git a/tests/cpp/util/CMakeLists.txt b/tests/cpp/util/CMakeLists.txt
index 1dfd2fed4e4..b80a588fe35 100644
--- a/tests/cpp/util/CMakeLists.txt
+++ b/tests/cpp/util/CMakeLists.txt
@@ -21,4 +21,6 @@ target_link_libraries(test_util PUBLIC
 target_compile_options(test_util PRIVATE -O2 -fopenmp)
 
 include(GoogleTest)
-gtest_discover_tests(test_util DISCOVERY_TIMEOUT 600)
+# MALLOC_PERTURB_ makes glibc overwrite freed memory, so a destructor that reads a destroyed object
+# at exit crashes the test process, which ctest reports as a failure, instead of passing silently.
+gtest_discover_tests(test_util DISCOVERY_TIMEOUT 600 PROPERTIES ENVIRONMENT "MALLOC_PERTURB_=165")
diff --git a/transformer_engine/common/fused_attn/graph_cache.h b/transformer_engine/common/fused_attn/graph_cache.h
index b3020f25acf..d52a0daf70b 100644
--- a/transformer_engine/common/fused_attn/graph_cache.h
+++ b/transformer_engine/common/fused_attn/graph_cache.h
@@ -130,7 +130,9 @@ std::shared_ptr<CacheEntry<GraphAndTensors>> cache_graph(GraphCache<GraphAndTens
 // Create a cache for each (backend, pass) pair, and either get cached entry or build anew.
 template <Backend kBackend, Pass kPass, auto kCreateGraphFn>
 auto get_graph(const FusedAttnConfig &cfg, cudnnHandle_t handle) {
-  static GraphCache<decltype(kCreateGraphFn(cfg))> cache;
+  // Never destroyed: destroying cached cuDNN plans calls into CUDA, which must not be used during
+  // program termination.
+  static auto &cache = *new GraphCache<decltype(kCreateGraphFn(cfg))>;
   cfg.check_derived();
   return detail::cache_graph(cache, cfg.make_cache_key(kPass), kBackend, kPass, handle,
                              [&] { return kCreateGraphFn(cfg); });
diff --git a/transformer_engine/common/util/cuda_driver.h b/transformer_engine/common/util/cuda_driver.h
index 16242347f12..3b85bd7cce4 100644
--- a/transformer_engine/common/util/cuda_driver.h
+++ b/transformer_engine/common/util/cuda_driver.h
@@ -40,8 +40,10 @@ template <typename... ArgTs>
 inline CUresult call(const char *symbol, ArgTs... args) {
   using FuncT = CUresult(ArgTs...);
 
-  static std::unordered_map<std::string, void *> symbol_cache;
-  static std::mutex cache_mutex;
+  // Never destroyed, so a lookup never reads a destroyed map, whatever order statics are
+  // constructed and destroyed in.
+  static auto &symbol_cache = *new std::unordered_map<std::string, void *>();
+  static auto &cache_mutex = *new std::mutex();
   FuncT *func;
 
   {
diff --git a/transformer_engine/common/util/rtc.cpp b/transformer_engine/common/util/rtc.cpp
index 2ffbbf72334..f591620124c 100644
--- a/transformer_engine/common/util/rtc.cpp
+++ b/transformer_engine/common/util/rtc.cpp
@@ -153,7 +153,9 @@ int Kernel::occupancy_max_active_blocks_per_sm(int device_id, int block_size,
 
 KernelManager& KernelManager::instance() {
   NVTE_CHECK(is_enabled(), "NVRTC support is not enabled");
-  static KernelManager instance_;
+  // Never destroyed: ~Kernel unloads modules through the CUDA driver, and CUDA must not be used
+  // during program termination.
+  static KernelManager& instance_ = *new KernelManager;
   return instance_;
 }