refactor <cuda/std/cstring>

NVIDIA · Jan 22, 2025 · 3793a2d · 3793a2d
1 parent 6a0f48b
commit 3793a2d
Show file tree

Hide file tree

Showing 481 changed files with 11,206 additions and 5,285 deletions.
diff --git a/.clangd b/.clangd
@@ -51,6 +51,7 @@ CompileFlags:
     # strip CUDA flags unknown to clang
     - "-ccbin*"
     - "--compiler-options*"
+    - "--extended-lambda"
     - "--expt-extended-lambda"
     - "--expt-relaxed-constexpr"
     - "-forward-unknown-to-host-compiler"

diff --git a/.github/CODEOWNERS b/.github/CODEOWNERS
@@ -1,20 +1,29 @@
-# general codeowners for all files
-# (Order matters. This needs to be at the top)
-* @nvidia/cccl-codeowners
-
 # Libraries
-thrust/ @nvidia/cccl-thrust-codeowners @nvidia/cccl-codeowners
-cub/ @nvidia/cccl-cub-codeowners @nvidia/cccl-codeowners
-libcudacxx/ @nvidia/cccl-libcudacxx-codeowners @nvidia/cccl-codeowners
+thrust/ @nvidia/cccl-thrust-codeowners
+cub/ @nvidia/cccl-cub-codeowners
+libcudacxx/ @nvidia/cccl-libcudacxx-codeowners
 cudax/ @nvidia/cccl-cudax-codeowners
 c/ @nvidia/cccl-c-codeowners
 python/ @nvidia/cccl-python-codeowners
 
 # Infrastructure
-.github/ @nvidia/cccl-infra-codeowners @nvidia/cccl-codeowners
-ci/ @nvidia/cccl-infra-codeowners @nvidia/cccl-codeowners
-.devcontainer/ @nvidia/cccl-infra-codeowners @nvidia/cccl-codeowners
+.github/ @nvidia/cccl-infra-codeowners
+ci/ @nvidia/cccl-infra-codeowners
+.devcontainer/ @nvidia/cccl-infra-codeowners
+.pre-commit-config.yaml @nvidia/cccl-infra-codeowners
+.clang-format @nvidia/cccl-infra-codeowners
+.clangd @nvidia/cccl-infra-codeowners
+c2h/ @nvidia/cccl-infra-codeowners
+.vscode @nvidia/cccl-infra-codeowners
 
 # cmake
-**/CMakeLists.txt @nvidia/cccl-cmake-codeowners @nvidia/cccl-codeowners
-**/cmake/ @nvidia/cccl-cmake-codeowners  @nvidia/cccl-codeowners
+**/CMakeLists.txt @nvidia/cccl-cmake-codeowners
+**/cmake/ @nvidia/cccl-cmake-codeowners
+
+# benchmarks
+benchmarks/ @nvidia/cccl-benchmark-codeowners
+**/benchmarks @nvidia/cccl-benchmark-codeowners
+
+# docs
+docs/ @nvidia/cccl-docs-codeowners
+examples/ @nvidia/cccl-docs-codeowners
diff --git a/.github/actions/docs-build/action.yml b/.github/actions/docs-build/action.yml
@@ -54,4 +54,4 @@ runs:
     # Upload docs as pages artifacts
     - name: Upload artifact
       if: ${{ inputs.upload_pages_artifact == 'true' }}
-      uses: actions/upload-pages-artifact@v2
+      uses: actions/upload-pages-artifact@v3
diff --git a/.github/workflows/build-docs.yml b/.github/workflows/build-docs.yml
@@ -45,4 +45,4 @@ jobs:
     steps:
       - name: Deploy to GitHub Pages
         id: deployment
-        uses: actions/deploy-pages@v2
+        uses: actions/deploy-pages@v4
diff --git a/.github/workflows/build-rapids.yml b/.github/workflows/build-rapids.yml
@@ -134,6 +134,12 @@ jobs:
               sccache --show-adv-stats
             done
           done
+
+          # Exit with error if any failures occurred
+          if test ${#failures[@]} -ne 0; then
+            exit 1
+          fi
+
           EOF
 
           chmod +x "$RUNNER_TEMP"/ci{,-entrypoint}.sh

diff --git a/.pre-commit-config.yaml b/.pre-commit-config.yaml
@@ -43,6 +43,17 @@ repos:
     hooks:
     - id: ruff  # linter
     - id: ruff-format  # formatter
+
+  # TOML lint & format
+  - repo: https://github.com/ComPWA/taplo-pre-commit
+    rev: v0.9.3
+    hooks:
+      # See https://github.com/NVIDIA/cccl/issues/3426
+      # - id: taplo-lint
+      #   exclude: "^docs/"
+      - id: taplo-format
+        exclude: "^docs/"
+
   - repo: https://github.com/codespell-project/codespell
     rev: v2.3.0
     hooks:

diff --git a/CMakePresets.json b/CMakePresets.json
@@ -73,8 +73,6 @@
         "CUB_ENABLE_DIALECT_CPP20": true,
         "THRUST_ENABLE_MULTICONFIG": true,
         "THRUST_MULTICONFIG_WORKLOAD": "LARGE",
-        "THRUST_MULTICONFIG_ENABLE_DIALECT_CPP11": true,
-        "THRUST_MULTICONFIG_ENABLE_DIALECT_CPP14": true,
         "THRUST_MULTICONFIG_ENABLE_DIALECT_CPP17": true,
         "THRUST_MULTICONFIG_ENABLE_DIALECT_CPP20": true,
         "THRUST_MULTICONFIG_ENABLE_SYSTEM_CPP": true,
@@ -128,28 +126,6 @@
         "LIBCUDACXX_ENABLE_LIBCUDACXX_TESTS": true
       }
     },
-    {
-      "name": "libcudacxx-cpp11",
-      "displayName": "libcu++: C++11",
-      "inherits": "libcudacxx-base",
-      "cacheVariables": {
-        "CMAKE_CXX_STANDARD": "11",
-        "CMAKE_CUDA_STANDARD": "11",
-        "LIBCUDACXX_TEST_STANDARD_VER": "c++11",
-        "CCCL_IGNORE_DEPRECATED_CPP_11": true
-      }
-    },
-    {
-      "name": "libcudacxx-cpp14",
-      "displayName": "libcu++: C++14",
-      "inherits": "libcudacxx-base",
-      "cacheVariables": {
-        "CMAKE_CXX_STANDARD": "14",
-        "CMAKE_CUDA_STANDARD": "14",
-        "LIBCUDACXX_TEST_STANDARD_VER": "c++14",
-        "CCCL_IGNORE_DEPRECATED_CPP_14": true
-      }
-    },
     {
       "name": "libcudacxx-cpp17",
       "displayName": "libcu++: C++17",
@@ -179,28 +155,6 @@
         "CMAKE_CUDA_ARCHITECTURES": "70"
       }
     },
-    {
-      "name": "libcudacxx-nvrtc-cpp11",
-      "displayName": "libcu++ NVRTC: C++11",
-      "inherits": "libcudacxx-nvrtc-base",
-      "cacheVariables": {
-        "CMAKE_CXX_STANDARD": "11",
-        "CMAKE_CUDA_STANDARD": "11",
-        "LIBCUDACXX_TEST_STANDARD_VER": "c++11",
-        "CCCL_IGNORE_DEPRECATED_CPP_11": true
-      }
-    },
-    {
-      "name": "libcudacxx-nvrtc-cpp14",
-      "displayName": "libcu++ NVRTC: C++14",
-      "inherits": "libcudacxx-nvrtc-base",
-      "cacheVariables": {
-        "CMAKE_CXX_STANDARD": "14",
-        "CMAKE_CUDA_STANDARD": "14",
-        "LIBCUDACXX_TEST_STANDARD_VER": "c++14",
-        "CCCL_IGNORE_DEPRECATED_CPP_14": true
-      }
-    },
     {
       "name": "libcudacxx-nvrtc-cpp17",
       "displayName": "libcu++ NVRTC: C++17",
@@ -261,8 +215,6 @@
         "THRUST_MULTICONFIG_ENABLE_SYSTEM_CUDA": true,
         "THRUST_MULTICONFIG_ENABLE_SYSTEM_OMP": true,
         "THRUST_MULTICONFIG_ENABLE_SYSTEM_TBB": true,
-        "THRUST_MULTICONFIG_ENABLE_DIALECT_CPP11": false,
-        "THRUST_MULTICONFIG_ENABLE_DIALECT_CPP14": false,
         "THRUST_MULTICONFIG_ENABLE_DIALECT_CPP17": false,
         "THRUST_MULTICONFIG_ENABLE_DIALECT_CPP20": false
       }
@@ -420,22 +372,6 @@
         "libcudacxx.test.atomics.ptx"
       ]
     },
-    {
-      "name": "libcudacxx-nvrtc-cpp11",
-      "hidden": false,
-      "inherits": [
-        "libcudacxx-nvrtcc"
-      ],
-      "configurePreset": "libcudacxx-nvrtc-cpp11"
-    },
-    {
-      "name": "libcudacxx-nvrtc-cpp14",
-      "hidden": false,
-      "inherits": [
-        "libcudacxx-nvrtcc"
-      ],
-      "configurePreset": "libcudacxx-nvrtc-cpp14"
-    },
     {
       "name": "libcudacxx-nvrtc-cpp17",
       "hidden": false,
@@ -452,20 +388,6 @@
       ],
       "configurePreset": "libcudacxx-nvrtc-cpp20"
     },
-    {
-      "name": "libcudacxx-cpp11",
-      "configurePreset": "libcudacxx-cpp11",
-      "inherits": [
-        "libcudacxx-base"
-      ]
-    },
-    {
-      "name": "libcudacxx-cpp14",
-      "configurePreset": "libcudacxx-cpp14",
-      "inherits": [
-        "libcudacxx-base"
-      ]
-    },
     {
       "name": "libcudacxx-cpp17",
       "configurePreset": "libcudacxx-cpp17",
@@ -572,20 +494,6 @@
         "outputOnFailure": false
       }
     },
-    {
-      "name": "libcudacxx-lit-cpp11",
-      "configurePreset": "libcudacxx-cpp11",
-      "inherits": [
-        "libcudacxx-lit-base"
-      ]
-    },
-    {
-      "name": "libcudacxx-lit-cpp14",
-      "configurePreset": "libcudacxx-cpp14",
-      "inherits": [
-        "libcudacxx-lit-base"
-      ]
-    },
     {
       "name": "libcudacxx-lit-cpp17",
       "configurePreset": "libcudacxx-cpp17",
@@ -607,20 +515,6 @@
         "libcudacxx-lit-base"
       ]
     },
-    {
-      "name": "libcudacxx-nvrtc-cpp11",
-      "configurePreset": "libcudacxx-nvrtc-cpp11",
-      "inherits": [
-        "libcudacxx-nvrtc-base"
-      ]
-    },
-    {
-      "name": "libcudacxx-nvrtc-cpp14",
-      "configurePreset": "libcudacxx-nvrtc-cpp14",
-      "inherits": [
-        "libcudacxx-nvrtc-base"
-      ]
-    },
     {
       "name": "libcudacxx-nvrtc-cpp17",
       "configurePreset": "libcudacxx-nvrtc-cpp17",

diff --git a/c/parallel/src/reduce.cu b/c/parallel/src/reduce.cu
@@ -160,7 +160,7 @@ std::string get_single_tile_kernel_name(
   check(nvrtcGetTypeName<op_wrapper>(&reduction_op_t));
 
   return std::format(
-    "cub::DeviceReduceSingleTileKernel<{0}, {1}, {2}, {3}, {4}, {5}, {6}>",
+    "cub::detail::reduce::DeviceReduceSingleTileKernel<{0}, {1}, {2}, {3}, {4}, {5}, {6}>",
     chained_policy_t,
     input_iterator_t,
     output_iterator_t,
@@ -192,7 +192,7 @@ std::string get_device_reduce_kernel_name(cccl_op_t op, cccl_iterator_t input_it
   check(nvrtcGetTypeName<cuda::std::__identity>(&transform_op_t));
 
   return std::format(
-    "cub::DeviceReduceKernel<{0}, {1}, {2}, {3}, {4}, {5}>",
+    "cub::detail::reduce::DeviceReduceKernel<{0}, {1}, {2}, {3}, {4}, {5}>",
     chained_policy_t,
     input_iterator_t,
     offset_t,

diff --git a/c/parallel/test/test_main.cpp b/c/parallel/test/test_main.cpp
@@ -12,8 +12,7 @@
 
 #include <iostream>
 
-#define CATCH_CONFIG_RUNNER
-#include <catch2/catch.hpp>
+#include <catch2/catch_session.hpp>
 
 int device_guard(int device_id)
 {
@@ -40,7 +39,7 @@ int main(int argc, char* argv[])
   int device_id{};
 
   // Build a new parser on top of Catch's
-  using namespace Catch::clara;
+  using namespace Catch::Clara;
   auto cli = session.cli() | Opt(device_id, "device")["-d"]["--device"]("device id to use");
   session.cli(cli);
 

diff --git a/c/parallel/test/test_util.h b/c/parallel/test/test_util.h
@@ -22,7 +22,9 @@
 #include <type_traits>
 #include <vector>
 
-#include <catch2/catch.hpp>
+#include <catch2/catch_template_test_macros.hpp>
+#include <catch2/catch_test_macros.hpp>
+#include <catch2/generators/catch_generators_all.hpp>
 #include <cccl/c/reduce.h>
 #include <nvrtc.h>
 

diff --git a/c2h/include/c2h/catch2_main.h b/c2h/include/c2h/catch2_main.h
@@ -36,13 +36,9 @@
 //! executable, this header is included into each test. On the other hand, when all the tests are compiled into a single
 //! executable, this header is excluded from the tests and included into catch2_runner.cpp
 
-#ifdef CUB_CONFIG_MAIN
-#  define CATCH_CONFIG_RUNNER
-#endif
-
-#include <catch2/catch.hpp>
+#include <catch2/catch_session.hpp>
 
-#if defined(CUB_CONFIG_MAIN)
+#ifdef CUB_CONFIG_MAIN
 #  if THRUST_DEVICE_SYSTEM == THRUST_DEVICE_SYSTEM_CUDA
 #    include <c2h/catch2_runner_helper.h>
 
@@ -59,7 +55,7 @@ int main(int argc, char* argv[])
   int device_id{};
 
   // Build a new parser on top of Catch's
-  using namespace Catch::clara;
+  using namespace Catch::Clara;
   auto cli = session.cli() | Opt(device_id, "device")["-d"]["--device"]("device id to use");
   session.cli(cli);
 
@@ -73,4 +69,4 @@ int main(int argc, char* argv[])
 #  endif // THRUST_DEVICE_SYSTEM == THRUST_DEVICE_SYSTEM_CUDA
   return session.run(argc, argv);
 }
-#endif
+#endif // CUB_CONFIG_MAIN