Develop stream 2025-01-20

.gitlab-ci.yml

@@ -1,5 +1,5 @@
 # ########################################################################
-# Copyright 2019-2024 Advanced Micro Devices, Inc.
+# Copyright 2019-2025 Advanced Micro Devices, Inc.
 # ########################################################################
 
 include:
@@ -23,6 +23,10 @@ stages:
   - test      # Tests if unit tests are passing (CTest)
   - benchmark # Runs the non-internal benchmarks (Google Benchmark)
 
+workflow:
+  rules:
+    - if: $CI_MERGE_REQUEST_LABELS !~ /CI Skip/
+
 variables:
   # Helper variables
   PACKAGE_DIR: $BUILD_DIR/package
@@ -74,7 +78,6 @@ copyright-date:
       -D CMAKE_CXX_COMPILER=hipcc
       -D CMAKE_BUILD_TYPE=Release
       -D BUILD_TEST=OFF
-      -D BUILD_HIPSTDPAR_TEST=OFF
       -D BUILD_EXAMPLE=OFF
       -D ROCM_DEP_ROCMCORE=OFF
       -D CMAKE_C_COMPILER_LAUNCHER=phc_sccache_c
@@ -116,7 +119,6 @@ copyright-date:
       -D CMAKE_BUILD_TYPE=$BUILD_TYPE
       -D BUILD_$BUILD_TARGET=ON
       -D GPU_TARGETS=$GPU_TARGETS
-      -D AMDGPU_TEST_TARGETS=$GPU_TARGETS
       -D RNG_SEED_COUNT=$rng_seed_count
       -D PRNG_SEEDS=$prng_seeds
       -D CMAKE_C_COMPILER_LAUNCHER=phc_sccache_c
@@ -137,7 +139,7 @@ copyright-date:
     - $CI_PROJECT_DIR/build/.ninja_log
     exclude:
     - $CI_PROJECT_DIR/build/**/*.o
-    expire_in: 2 weeks
+    expire_in: 1 day
 
 build:cmake-latest:
   stage: build
@@ -183,7 +185,7 @@ build:package:
     paths:
       - $PACKAGE_DIR/rocthrust*.deb
       - $PACKAGE_DIR/rocthrust*.zip
-    expire_in: 2 weeks
+    expire_in: 1 day
 
 build:windows:
   stage: build
@@ -231,7 +233,7 @@ build:windows:
     paths:
       - $CI_PROJECT_DIR/build/
       - $ROCPRIM_DIR/build/install
-    expire_in: 2 weeks
+    expire_in: 1 day
 
 test:package:
   stage: test
@@ -268,13 +270,18 @@ test:doc:
   extends:
     - .build:docs
     - .rules:test
+  artifacts:
+    paths:
+      - $DOCS_DIR/_build/html/
+    expire_in: 2 weeks
 
-test:
+.test:rocm:
   stage: test
+  tags:
+    - rocm
+    - $GPU
   extends:
     - .cmake-minimum
-    - .rules:test
-    - .gpus:rocm
   needs:
     - job: build:cmake-minimum
       parallel:
@@ -294,10 +301,76 @@ test:
     - HSA_ENABLE_SDMA=0 ctest
       --output-on-failure
       --repeat-until-fail 2
-      --tests-regex $GPU_TARGET
       --resource-spec-file ./resources.json
       --parallel $PARALLEL_JOBS
 
+test:rocm-any-gpu:
+  variables:
+    GPU: ""
+    PARALLEL_JOBS: 1
+  extends:
+    - .test:rocm
+  rules:
+    - if: $CI_MERGE_REQUEST_TITLE =~ /Draft:/ && $CI_MERGE_REQUEST_LABELS !~ /Arch::/
+
+test:rocm-label-arch:
+  extends:
+    - .gpus:rocm
+    - .test:rocm
+    - .rules:arch-labels
+
+test:rocm-all-gpus:
+  variables:
+    SHOULD_BE_UNDRAFTED: "true"
+  extends:
+    - .gpus:rocm
+    - .test:rocm
+    - .rules:test
+
+test-bitwise-repro-generate:
+  stage: test
+  extends:
+    - .cmake-minimum
+    - .rules:manual
+    - .gpus:rocm
+  needs:
+    - job: build:cmake-minimum
+      parallel:
+        matrix:
+          - BUILD_TYPE: Release
+            BUILD_TARGET: TEST
+            BUILD_VERSION: 14
+  cache:
+    key: database
+    paths:
+      - $CI_PROJECT_DIR/bitwise.db
+    policy: push
+  script:
+    - cd $CI_PROJECT_DIR/build/test/
+    - ROCTHRUST_BWR_PATH=$CI_PROJECT_DIR/bitwise.db ROCTHRUST_BWR_GENERATE=1 ./reproducibility.hip
+
+test-bitwise-repro:
+  stage: test
+  extends:
+    - .cmake-minimum
+    - .rules:test
+    - .gpus:rocm
+  needs:
+    - job: build:cmake-minimum
+      parallel:
+        matrix:
+          - BUILD_TYPE: Release
+            BUILD_TARGET: TEST
+            BUILD_VERSION: 14
+  cache:
+    key: database
+    paths:
+      - $CI_PROJECT_DIR/bitwise.db
+    policy: pull
+  script:
+    - cd $CI_PROJECT_DIR/build/test/
+    - ROCTHRUST_BWR_PATH=$CI_PROJECT_DIR/bitwise.db ./reproducibility.hip
+
 .rocm-windows:test:
   extends:
     - .gpus:rocm-windows
@@ -358,7 +431,7 @@ build:cuda-and-omp:
   tags:
     - build
   variables:
-    CCCL_GIT_BRANCH: v2.5.0
+    CCCL_GIT_BRANCH: v2.6.0
     CCCL_DIR: ${CI_PROJECT_DIR}/cccl
   needs: []
   script:
@@ -402,16 +475,17 @@ build:cuda-and-omp:
       - $CCCL_DIR/thrust/cmake/ThrustRunTest.cmake
       - $CCCL_DIR/thrust/cmake/ThrustRunExample.cmake
       - $CI_PROJECT_DIR/build/.ninja_log
-    expire_in: 1 week
+    expire_in: 1 day
 
-test:cuda-and-omp:
+.test:cuda-and-omp:
   stage: test
+  tags:
+    - nvcc
+    - $GPU
   needs:
     - build:cuda-and-omp
   extends:
     - .nvcc
-    - .gpus:nvcc
-    - .rules:test
   before_script:
     # This is only needed because of the legacy before_script in .gpus:nvcc would otherwise overwrite before_script
     - !reference [.nvcc, before_script]
@@ -420,6 +494,23 @@ test:cuda-and-omp:
     # These tests are executed on the build stage because they require sources
     - ctest --output-on-failure --exclude-regex "thrust.example.cmake.add_subdir|thrust.test.cmake.check_source_files"
 
+test:cuda-and-omp-any-gpu:
+  variables:
+    GPU: ""
+    PARALLEL_JOBS: 1
+  extends:
+    - .test:cuda-and-omp
+  rules:
+    - if: $CI_MERGE_REQUEST_TITLE =~ /Draft:/
+
+test:cuda-and-omp-all-gpus:
+  variables:
+    SHOULD_BE_UNDRAFTED: "true"
+  extends:
+    - .gpus:nvcc
+    - .test:cuda-and-omp
+    - .rules:test
+
 .benchmark-base:
   stage: benchmark
   extends:
@@ -430,7 +521,12 @@ test:cuda-and-omp:
 
 benchmark:
   needs:
-    - build:cmake-minimum
+    - job: build:cmake-minimum
+      parallel:
+        matrix:
+          - BUILD_TYPE: Release
+            BUILD_TARGET: BENCHMARKS
+            BUILD_VERSION: 14
   extends:
     - .cmake-minimum
     - .gpus:rocm
@@ -453,4 +549,4 @@ benchmark:
   artifacts:
     paths:
       - ${BENCHMARK_RESULT_DIR}
-    expire_in: 1 week
+    expire_in: 1 day

CMakeLists.txt

@@ -1,5 +1,5 @@
 # ########################################################################
-# Copyright 2019-2024 Advanced Micro Devices, Inc.
+# Copyright 2019-2025 Advanced Micro Devices, Inc.
 # ########################################################################
 
 cmake_minimum_required(VERSION 3.10.2 FATAL_ERROR)
@@ -84,6 +84,7 @@ option(BUILD_HIPSTDPAR_TEST "Build hipstdpar tests" OFF)
 option(BUILD_EXAMPLES "Build examples" OFF)
 option(BUILD_BENCHMARKS "Build benchmarks" OFF)
 option(DOWNLOAD_ROCPRIM "Download rocPRIM and do not search for rocPRIM package" OFF)
+cmake_dependent_option(DOWNLOAD_ROCRAND "Download rocRAND and do not search for rocRAND package" OFF BUILD_BENCHMARKS OFF)
 option(BUILD_ADDRESS_SANITIZER "Build with address sanitizer enabled" OFF)
 cmake_dependent_option(ENABLE_UPSTREAM_TESTS "Enable upstream (thrust) tests" ON BUILD_TEST OFF)
 #Set the header wrapper OFF by default.
@@ -113,10 +114,9 @@ endif()
 set(CMAKE_CXX_STANDARD_REQUIRED ON)
 set(CMAKE_CXX_EXTENSIONS OFF)
 
-if(DISABLE_WERROR)
-  set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -Wall -Wextra ")
-else()
-  set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -Wall -Wextra -Werror")
+set(COMPILE_OPTIONS -Wall -Wextra)
+if(NOT DISABLE_WERROR)
+  list(APPEND COMPILE_OPTIONS -Werror)
 endif()
 
 if (CMAKE_CXX_STANDARD EQUAL 14)
@@ -132,13 +132,13 @@ endif()
 
 # Address Sanitizer
 if(BUILD_ADDRESS_SANITIZER)
-  set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -fsanitize=address -shared-libasan")
-  set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -fsanitize=address -shared-libasan")
+  list(APPEND COMPILE_OPTIONS -fsanitize=address -shared-libasan)
   add_link_options(-fuse-ld=lld)
 endif()
 
 # Setup VERSION
 rocm_setup_version(VERSION "3.3.0")
+math(EXPR rocthrust_VERSION_NUMBER "${rocthrust_VERSION_MAJOR} * 100000 + ${rocthrust_VERSION_MINOR} * 100 + ${rocthrust_VERSION_PATCH}")
 
 # Print configuration summary
 include(cmake/Summary.cmake)
@@ -174,7 +174,6 @@ endif()
 # Benchmarks
 if(BUILD_BENCHMARKS)
   add_subdirectory(benchmarks)
-  add_subdirectory(internal/benchmark)
 endif()
 
 #Create header wrapper for backward compatibility

README.md

@@ -1,7 +1,7 @@
 # rocThrust
 
 > [!NOTE]
-> The published documentation is available at [rocThrust](https://rocm.docs.amd.com/projects/rocThrust/en/latest/) in an organized, easy-to-read format, with search and a table of contents. The documentation source files reside in the `docs` folder of this repository. As with all ROCm projects, the documentation is open source. For more information on contributing to the documentation, see [Contribute to ROCm documentation](https://rocm.docs.amd.com/en/latest/contribute/contributing.html).
+> The published rocThrust documentation is available [here](https://rocm.docs.amd.com/projects/rocThrust/en/latest/) in an organized, easy-to-read format, with search and a table of contents. The documentation source files reside in the `docs` folder of this repository. As with all ROCm projects, the documentation is open source. For more information on contributing to the documentation, see [Contribute to ROCm documentation](https://rocm.docs.amd.com/en/latest/contribute/contributing.html).
 
 Thrust is a parallel algorithm library. It has been ported to
 [HIP](https://github.com/ROCm/HIP) and [ROCm](https://www.github.com/ROCm/ROCm), which use
@@ -35,28 +35,6 @@ For ROCm hardware requirements, refer to:
 * [Linux support](https://rocm.docs.amd.com/projects/install-on-linux/en/latest/reference/system-requirements.html)
 * [Windows support](https://rocm.docs.amd.com/projects/install-on-windows/en/latest/reference/system-requirements.html)
 
-## Documentation
-
-Documentation for rocThrust available at
-[https://rocm.docs.amd.com/projects/rocThrust/en/latest/](https://rocm.docs.amd.com/projects/rocThrust/en/latest/).
-
-You can build our documentation locally using the following commands:
-
-```shell
-# Go to rocThrust docs directory
-cd rocThrust; cd docs
-
-# Install Python dependencies
-python3 -m pip install -r sphinx/requirements.txt
-
-# Build the documentation
-python3 -m sphinx -T -E -b html -d _build/doctrees -D language=en . _build/html
-
-# For e.g. serve the HTML docs locally
-cd _build/html
-python3 -m http.server
-```
-
 ## Build and install
 
 ```sh
@@ -309,6 +287,58 @@ ctest --output-on-failure
   * Notice that oneTBB (oneAPI TBB) may fail to compile when libstdc++-9 or -10 is used, due to them using legacy TBB interfaces that are incompatible with the oneTBB ones (see the [release notes](https://www.intel.com/content/www/us/en/developer/articles/release-notes/intel-oneapi-threading-building-blocks-release-notes.html)).
 * CMake (3.10.2 or later)
 
+## Building the documentation locally
+
+### Requirements
+
+#### Doxygen
+
+The build system uses Doxygen [version 1.9.4](https://github.com/doxygen/doxygen/releases/tag/Release_1_9_4). You can try using a newer version, but that might cause issues.
+
+After you have downloaded Doxygen version 1.9.4:
+
+```shell
+# Add doxygen to your PATH
+echo 'export PATH=<doxygen 1.9.4 path>/bin:$PATH' >> ~/.bashrc
+
+# Apply the updated .bashrc
+source ~/.bashrc
+
+# Confirm that you are using version 1.9.4
+doxygen --version
+```
+
+#### Python
+
+The build system uses Python version 3.10. You can try using a newer version, but that might cause issues.
+
+You can install Python 3.10 alongside your other Python versions using [pyenv](https://github.com/pyenv/pyenv?tab=readme-ov-file#installation):
+
+```shell
+# Install Python 3.10
+pyenv install 3.10
+
+# Create a Python 3.10 virtual environment
+pyenv virtualenv 3.10 venv_rocthrust
+
+# Activate the virtual environment
+pyenv activate venv_rocthrust
+```
+
+### Building
+
+After cloning this repository, and `cd`ing into it:
+
+```shell
+# Install Python dependencies
+python3 -m pip install -r docs/sphinx/requirements.txt
+
+# Build the documentation
+python3 -m sphinx -T -E -b html -d docs/_build/doctrees -D language=en docs docs/_build/html
+```
+
+You can then open `docs/_build/html/index.html` in your browser to view the documentation.
+
 ## Support
 
 You can report bugs and feature requests through the GitHub