Skip async mr tests when cuda runtime/driver < 11.2

ci/gpu/build.sh

@@ -102,7 +102,7 @@ else
 
     gpuci_logger "Running googletests"
     # run gtests from librmm_tests package
-    for gt in "$CONDA_PREFIX/bin/gtests/librmm/*" ; do
+    for gt in "$CONDA_PREFIX/bin/gtests/librmm/"* ; do
         ${gt} --gtest_output=xml:${TESTRESULTS_DIR}/
         exitcode=$?
         if (( ${exitcode} != 0 )); then

include/rmm/mr/device/cuda_async_memory_resource.hpp

@@ -62,19 +62,15 @@ class cuda_async_memory_resource final : public device_memory_resource {
   {
 #ifdef RMM_CUDA_MALLOC_ASYNC_SUPPORT
     // Check if cudaMallocAsync Memory pool supported
-    auto const device = rmm::detail::current_device();
-    int cuda_pool_supported{};
-    auto result =
-      cudaDeviceGetAttribute(&cuda_pool_supported, cudaDevAttrMemoryPoolsSupported, device.value());
-    RMM_EXPECTS(result == cudaSuccess && cuda_pool_supported,
+    RMM_EXPECTS(is_supported(),
                 "cudaMallocAsync not supported with this CUDA driver/runtime version");
 
     // Construct explicit pool
     cudaMemPoolProps pool_props{};
     pool_props.allocType     = cudaMemAllocationTypePinned;
     pool_props.handleTypes   = cudaMemHandleTypePosixFileDescriptor;
     pool_props.location.type = cudaMemLocationTypeDevice;
-    pool_props.location.id   = device.value();
+    pool_props.location.id   = rmm::detail::current_device().value();
     RMM_CUDA_TRY(cudaMemPoolCreate(&cuda_pool_handle_, &pool_props));
 
     auto const [free, total] = rmm::detail::available_device_memory();
@@ -115,6 +111,32 @@ class cuda_async_memory_resource final : public device_memory_resource {
   cuda_async_memory_resource& operator=(cuda_async_memory_resource const&) = delete;
   cuda_async_memory_resource& operator=(cuda_async_memory_resource&&) = delete;
 
+  /**
+   * @brief Is cudaMallocAsync supported with this cuda runtime/driver version?
+   * @return true if both the cuda runtime and driver are newer than 11.2
+   */
+  static bool is_supported()
+  {
+#if defined(RMM_CUDA_MALLOC_ASYNC_SUPPORT)
+    static auto runtime_supports_pool{[] {
+      int runtime_version{};
+      RMM_CUDA_TRY(cudaRuntimeGetVersion(&runtime_version));
+      constexpr auto min_async_version{11020};
+      return runtime_version >= min_async_version;
+    }()};
+    static auto driver_supports_pool{[] {
+      int cuda_pool_supported{};
+      auto result = cudaDeviceGetAttribute(&cuda_pool_supported,
+                                           cudaDevAttrMemoryPoolsSupported,
+                                           rmm::detail::current_device().value());
+      return result == cudaSuccess and cuda_pool_supported == 1;
+    }()};
+    return runtime_supports_pool and driver_supports_pool;
+#else
+    return false;
+#endif
+  }
+
   /**
    * @brief Query whether the resource supports use of non-null CUDA streams for
    * allocation/deallocation. `cuda_memory_resource` does not support streams.

tests/mr/device/cuda_async_mr_tests.cpp

@@ -24,7 +24,18 @@ namespace {
 
 using cuda_async_mr = rmm::mr::cuda_async_memory_resource;
 
-TEST(AsyncMRTest, ThrowIfNotSupported)
+class AsyncMRTest : public ::testing::Test {
+ protected:
+  void SetUp() override
+  {
+    if (!rmm::mr::cuda_async_memory_resource::is_supported()) {
+      GTEST_SKIP() << "Skipping tests since cudaMallocAsync not supported with this CUDA "
+                   << "driver/runtime version";
+    }
+  }
+};
+
+TEST_F(AsyncMRTest, ThrowIfNotSupported)
 {
   auto construct_mr = []() { cuda_async_mr mr; };
 #ifndef RMM_CUDA_MALLOC_ASYNC_SUPPORT
@@ -35,7 +46,7 @@ TEST(AsyncMRTest, ThrowIfNotSupported)
 }
 
 #if defined(RMM_CUDA_MALLOC_ASYNC_SUPPORT)
-TEST(AsyncMRTest, ExplicitInitialPoolSize)
+TEST_F(AsyncMRTest, ExplicitInitialPoolSize)
 {
   const auto pool_init_size{100};
   cuda_async_mr mr{pool_init_size};
@@ -44,7 +55,7 @@ TEST(AsyncMRTest, ExplicitInitialPoolSize)
   RMM_CUDA_TRY(cudaDeviceSynchronize());
 }
 
-TEST(AsyncMRTest, ExplicitReleaseThreshold)
+TEST_F(AsyncMRTest, ExplicitReleaseThreshold)
 {
   const auto pool_init_size{100};
   const auto pool_release_threshold{1000};
@@ -54,7 +65,7 @@ TEST(AsyncMRTest, ExplicitReleaseThreshold)
   RMM_CUDA_TRY(cudaDeviceSynchronize());
 }
 
-TEST(AsyncMRTest, DifferentPoolsUnequal)
+TEST_F(AsyncMRTest, DifferentPoolsUnequal)
 {
   const auto pool_init_size{100};
   const auto pool_release_threshold{1000};

tests/mr/device/mr_test.hpp

@@ -234,6 +234,10 @@ struct mr_test : public ::testing::TestWithParam<mr_factory> {
   {
     auto factory = GetParam().factory;
     mr           = factory();
+    if (mr == nullptr) {
+      GTEST_SKIP() << "Skipping tests since the memory resource is not supported with this CUDA "
+                   << "driver/runtime version";
+    }
   }
 
   std::shared_ptr<rmm::mr::device_memory_resource> mr;  ///< Pointer to resource to use in tests
@@ -246,7 +250,13 @@ struct mr_allocation_test : public mr_test {
 /// MR factory functions
 inline auto make_cuda() { return std::make_shared<rmm::mr::cuda_memory_resource>(); }
 
-inline auto make_cuda_async() { return std::make_shared<rmm::mr::cuda_async_memory_resource>(); }
+inline auto make_cuda_async()
+{
+  if (rmm::mr::cuda_async_memory_resource::is_supported()) {
+    return std::make_shared<rmm::mr::cuda_async_memory_resource>();
+  }
+  return std::shared_ptr<rmm::mr::cuda_async_memory_resource>{nullptr};
+}
 
 inline auto make_managed() { return std::make_shared<rmm::mr::managed_memory_resource>(); }