Merge remote-tracking branch 'origin/main' into pr/allisonvacanti/300

.devcontainer/cuda12.2-llvm16/devcontainer.json

@@ -0,0 +1,39 @@
+{
+  "shutdownAction": "stopContainer",
+  "image": "rapidsai/devcontainers:23.08-cpp-llvm16-cuda12.2-ubuntu22.04",
+  "hostRequirements": {
+    "gpu": true
+  },
+  "initializeCommand": [
+    "/bin/bash",
+    "-c",
+    "mkdir -m 0755 -p ${localWorkspaceFolder}/.{aws,cache,config}"
+  ],
+  "containerEnv": {
+    "SCCACHE_REGION": "us-east-2",
+    "SCCACHE_BUCKET": "rapids-sccache-devs",
+    "VAULT_HOST": "https://vault.ops.k8s.rapids.ai",
+    "HISTFILE": "${containerWorkspaceFolder}/.cache/._bash_history",
+    "DEVCONTAINER_NAME": "cuda12.2-llvm16"
+  },
+  "workspaceFolder": "/home/coder/${localWorkspaceFolderBasename}",
+  "workspaceMount": "source=${localWorkspaceFolder},target=/home/coder/${localWorkspaceFolderBasename},type=bind,consistency=consistent",
+  "mounts": [
+    "source=${localWorkspaceFolder}/.aws,target=/home/coder/.aws,type=bind,consistency=consistent",
+    "source=${localWorkspaceFolder}/.cache,target=/home/coder/.cache,type=bind,consistency=consistent",
+    "source=${localWorkspaceFolder}/.config,target=/home/coder/.config,type=bind,consistency=consistent"
+  ],
+  "customizations": {
+    "vscode": {
+      "extensions": [
+        "llvm-vs-code-extensions.vscode-clangd"
+      ],
+      "settings": {
+        "clangd.arguments": [
+          "--compile-commands-dir=${workspaceFolder}/build/latest"
+        ]
+      }
+    }
+  },
+  "name": "cuda12.2-llvm16"
+}

.github/ISSUE_TEMPLATE/bug_report.yml

@@ -1,7 +1,7 @@
 name: Bug Report
 description: Create a report to help us improve
 title: '[BUG]: '
-labels: ['Bug']
+labels: ['bug']
 body:
   - type: checkboxes
     id: check-duplicates

.github/ISSUE_TEMPLATE/doc_request.yml

@@ -1,7 +1,7 @@
 name: Documentation Request
 description: Suggest an idea to improve CCCL
 title: '[DOC]: '
-labels: ['Doc']
+labels: ['doc']
 
 body:
   - type: checkboxes

.github/ISSUE_TEMPLATE/feature_request.yml

@@ -1,7 +1,7 @@
 name: Feature Request
 description: Suggest an idea to improve CCCL
 title: '[FEA]: '
-labels: ['Feature']
+labels: ['feature request']
 body:
   - type: checkboxes
     id: check-duplicates

.github/copy-pr-bot.yaml

@@ -0,0 +1,4 @@
+# Configuration file for `copy-pr-bot` GitHub App
+# https://docs.gha-runners.nvidia.com/apps/copy-pr-bot/
+
+enabled: true

.github/workflows/pr.yml

@@ -1,3 +1,17 @@
+# SPDX-FileCopyrightText: Copyright (c) 2023 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
+# SPDX-License-Identifier: Apache-2.0
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
 
 # This is the main workflow that runs on every PR and push to main
 name: pr
@@ -18,7 +32,7 @@ concurrency:
   cancel-in-progress: true
 
 jobs:
-  
+
   get-devcontainer-version:
     name: Get devcontainer version
     runs-on: ubuntu-latest
@@ -37,6 +51,7 @@ jobs:
     name: Compute NVCC matrix
     runs-on: ubuntu-latest
     outputs:
+      FULL_MATRIX: ${{ steps.set-outputs.outputs.FULL_MATRIX }}
       CUDA_VERSIONS: ${{ steps.set-outputs.outputs.CUDA_VERSIONS }}
       HOST_COMPILERS: ${{ steps.set-outputs.outputs.HOST_COMPILERS }}
       PER_CUDA_COMPILER_MATRIX: ${{ steps.set-outputs.outputs.PER_CUDA_COMPILER_MATRIX }}
@@ -53,21 +68,59 @@ jobs:
         id: set-outputs
         run: |
           FULL_MATRIX='${{steps.compute-nvcc-matrix.outputs.matrix}}'
+          echo "FULL_MATRIX=$FULL_MATRIX" | tee -a "$GITHUB_OUTPUT"
           CUDA_VERSIONS=$(echo $FULL_MATRIX | jq -c '[.[] | .cuda] | unique')
           echo "CUDA_VERSIONS=$CUDA_VERSIONS" | tee -a "$GITHUB_OUTPUT"
           HOST_COMPILERS=$(echo $FULL_MATRIX | jq -c '[.[] | .compiler.name] | unique')
           echo "HOST_COMPILERS=$HOST_COMPILERS" | tee -a "$GITHUB_OUTPUT"
           PER_CUDA_COMPILER_MATRIX=$(echo $FULL_MATRIX | jq -c ' group_by(.cuda + .compiler.name) | map({(.[0].cuda + "-" + .[0].compiler.name): .}) | add')
           echo "PER_CUDA_COMPILER_MATRIX=$PER_CUDA_COMPILER_MATRIX" | tee -a "$GITHUB_OUTPUT"
 
+  compute-nvrtc-matrix:
+    name: Compute NVRTC matrix
+    runs-on: ubuntu-latest
+    outputs:
+      NVRTC_MATRIX: ${{ steps.set-outputs.outputs.NVRTC_MATRIX }}
+    steps:
+      - name: Checkout repo
+        uses: actions/checkout@v3
+      - name: Get full nvrtc matrix
+        id: compute-nvrtc-matrix
+        uses: ./.github/actions/compute-matrix
+        with:
+          matrix_file: './ci/matrix.yaml'
+          matrix_query: '.pull_request.nvrtc'
+      - name: Set outputs
+        id: set-outputs
+        run: |
+          NVRTC_MATRIX='${{steps.compute-nvrtc-matrix.outputs.matrix}}'
+          echo "NVRTC_MATRIX=$NVRTC_MATRIX" | tee -a "$GITHUB_OUTPUT"
+
+  nvrtc:
+    name: NVRTC CUDA${{matrix.cuda}} C++${{matrix.std}}
+    if: ${{ !contains(github.event.head_commit.message, 'skip-tests') }}
+    needs: [compute-nvrtc-matrix, get-devcontainer-version]
+    uses: ./.github/workflows/run-as-coder.yml
+    strategy:
+      fail-fast: false
+      matrix:
+        include: ${{ fromJSON(needs.compute-nvrtc-matrix.outputs.NVRTC_MATRIX) }}
+    with:
+      name: NVRTC CUDA${{matrix.cuda}} C++${{matrix.std}}
+      runner: linux-${{matrix.cpu}}-gpu-v100-latest-1
+      image: rapidsai/devcontainers:${{needs.get-devcontainer-version.outputs.DEVCONTAINER_VERSION}}-cpp-gcc12-cuda${{matrix.cuda}}-${{matrix.os}}
+      command: |
+        ./ci/nvrtc_libcudacxx.sh g++ ${{matrix.std}} ${{matrix.gpu_build_archs}}
+
+
   thrust:
     name: Thrust CUDA${{ matrix.cuda_version }} ${{ matrix.compiler }}
     needs: [compute-nvcc-matrix, get-devcontainer-version]
     uses: ./.github/workflows/dispatch-build-and-test.yml
     strategy:
       fail-fast: false
-      matrix: 
-        cuda_version: ${{ fromJSON(needs.compute-nvcc-matrix.outputs.CUDA_VERSIONS) }} 
+      matrix:
+        cuda_version: ${{ fromJSON(needs.compute-nvcc-matrix.outputs.CUDA_VERSIONS) }}
         compiler: ${{ fromJSON(needs.compute-nvcc-matrix.outputs.HOST_COMPILERS) }}
     with:
       per_cuda_compiler_matrix: ${{ toJSON(fromJSON(needs.compute-nvcc-matrix.outputs.PER_CUDA_COMPILER_MATRIX)[ format('{0}-{1}', matrix.cuda_version, matrix.compiler) ]) }}
@@ -81,38 +134,57 @@ jobs:
     uses: ./.github/workflows/dispatch-build-and-test.yml
     strategy:
       fail-fast: false
-      matrix: 
-        cuda_version: ${{ fromJSON(needs.compute-nvcc-matrix.outputs.CUDA_VERSIONS) }} 
+      matrix:
+        cuda_version: ${{ fromJSON(needs.compute-nvcc-matrix.outputs.CUDA_VERSIONS) }}
         compiler: ${{ fromJSON(needs.compute-nvcc-matrix.outputs.HOST_COMPILERS) }}
     with:
       per_cuda_compiler_matrix: ${{ toJSON(fromJSON(needs.compute-nvcc-matrix.outputs.PER_CUDA_COMPILER_MATRIX)[ format('{0}-{1}', matrix.cuda_version, matrix.compiler) ]) }}
       build_script: "./ci/build_cub.sh"
       test_script: "./ci/test_cub.sh"
       devcontainer_version: ${{ needs.get-devcontainer-version.outputs.DEVCONTAINER_VERSION }}
-  
+
   libcudacxx:
     name: libcudacxx CUDA${{ matrix.cuda_version }} ${{ matrix.compiler }}
     needs: [compute-nvcc-matrix, get-devcontainer-version]
     uses: ./.github/workflows/dispatch-build-and-test.yml
     strategy:
       fail-fast: false
-      matrix: 
-        cuda_version: ${{ fromJSON(needs.compute-nvcc-matrix.outputs.CUDA_VERSIONS) }} 
+      matrix:
+        cuda_version: ${{ fromJSON(needs.compute-nvcc-matrix.outputs.CUDA_VERSIONS) }}
         compiler: ${{ fromJSON(needs.compute-nvcc-matrix.outputs.HOST_COMPILERS) }}
     with:
       per_cuda_compiler_matrix: ${{ toJSON(fromJSON(needs.compute-nvcc-matrix.outputs.PER_CUDA_COMPILER_MATRIX)[ format('{0}-{1}', matrix.cuda_version, matrix.compiler) ]) }}
       build_script: "./ci/build_libcudacxx.sh"
-      test_script: "./ci/test_libcudacxx.sh" 
+      test_script: "./ci/test_libcudacxx.sh"
       devcontainer_version: ${{ needs.get-devcontainer-version.outputs.DEVCONTAINER_VERSION }}
 
+  examples:
+    name: CCCL Examples
+    if: ${{ !contains(github.event.head_commit.message, 'skip-tests') }}
+    needs: [compute-nvcc-matrix, get-devcontainer-version]
+    strategy:
+      fail-fast: false
+      matrix:
+        include: ${{ fromJSON(needs.compute-nvcc-matrix.outputs.FULL_MATRIX) }}
+    uses: ./.github/workflows/run-as-coder.yml
+    with:
+      name: CCCL Examples CUDA${{matrix.cuda}} ${{matrix.compiler.name}}${{matrix.compiler.version}}
+      runner: linux-${{matrix.cpu}}-gpu-v100-latest-1
+      image: rapidsai/devcontainers:${{needs.get-devcontainer-version.outputs.DEVCONTAINER_VERSION}}-cpp-${{matrix.compiler.name}}${{matrix.compiler.version}}-cuda${{matrix.cuda}}-${{matrix.os}}
+      command: |
+        cmake -S . --preset=examples -DCCCL_EXAMPLE_CPM_TAG=${GITHUB_SHA} -DCMAKE_CUDA_COMPILER=nvcc
+        ctest --preset=examples
+
   # This job is the final job that runs after all other jobs and is used for branch protection status checks.
   # See: https://docs.github.com/en/pull-requests/collaborating-with-pull-requests/collaborating-on-repositories-with-code-quality-features/about-status-checks
   ci:
     runs-on: ubuntu-latest
     name: CI
     needs:
-      - libcudacxx
       - cub
+      - libcudacxx
+      - nvrtc
       - thrust
+      - examples
     steps:
       - run: echo "CI success"

.github/workflows/project_automation_set_in_progress.yml

@@ -42,31 +42,49 @@ env:
 jobs:
   query_and_mutate_project_fields:
     runs-on: ubuntu-latest
+
+    permissions:
+      issues: write
+      pull-requests: write
 
     steps:
       - name: Check if changes requested from a reviewer
         id: check_changes_requested
+        if: github.event_name == 'pull_request_review'
         run: |
-          if [ ${{ github.event_name }} == 'pull_request_review' ]; then
-            if [ ${{ github.event.review.state }} != 'changes_requested' ]; then
-              echo "Changes not requested, exiting"
-              exit 0
-            fi
+          if [ ${{ github.event.review.state }} != 'changes_requested' ]; then
+            echo "Changes not requested, exiting"
+            exit 0
+          
+          # If it is requesting changes, set PR to draft
+          # We use the default token here since we're granting write access to the PR
+          elif [ ${{ github.event.pull_request.draft }} == true ]; then
+            gh api graphql -f query='
+              mutation {
+                convertPullRequestToDraft(input: {pullRequestId: "${{ env.PR_GLOBAL_ID }}"}) {
+                  clientMutationId
+                }
+              }'
+            exit 0
           fi
+        continue-on-error: true
 
       - name: Generate token
         id: generate_token
+        if: github.event_name == 'pull_request_target'
         uses: tibdex/github-app-token@v1.8.0
         with:
           app_id: ${{ secrets.CCCL_AUTH_APP_ID }}
           private_key: ${{ secrets.CCCL_AUTH_APP_PEM }}
 
       - name: Wait 1 Second
         id: sleep
+        if: github.event_name == 'pull_request_target'
         run: sleep 1
 
       - name: Get PR Project ID
         id: get_pr_id
+        if: github.event_name == 'pull_request_target'
         env:
           GITHUB_TOKEN: ${{ steps.generate_token.outputs.token }}
         run: |
@@ -104,6 +122,7 @@ jobs:
 
       - name: Set PR Fields
         id: set_pr_fields
+        if: github.event_name == 'pull_request_target'
         env:
           GITHUB_TOKEN: ${{ steps.generate_token.outputs.token }}
         run: |
@@ -233,6 +252,7 @@ jobs:
 
       - name: Sync Linked Issues
         id: sync_linked_issues
+        if: github.event_name == 'pull_request_target'
         env:
           GITHUB_TOKEN: ${{ steps.generate_token.outputs.token }}
         run: |
@@ -334,19 +354,3 @@ jobs:
                 fi
             done
         continue-on-error: true
-
-      - name: set PR to draft
-        id: set_pr_draft
-        # only run if PR is not a draft
-        if: ${{ github.event.pull_request.draft }} == false
-        env:
-            GITHUB_TOKEN: ${{ steps.generate_token.outputs.token }}
-        run: |
-            # Set the PR to draft using the graphQL API
-            gh api graphql -f query='
-              mutation {
-                convertPullRequestToDraft(input: {pullRequestId: "${{ env.PR_GLOBAL_ID }}"}) {
-                  clientMutationId
-                }
-              }'
-        continue-on-error: true

.github/workflows/run-as-coder.yml

@@ -40,6 +40,7 @@ jobs:
       - name: Run command
         shell: su coder {0}
         run: | 
+            set -exo pipefail
             cd ~/cccl
             eval "${{inputs.command}}" || exit_code=$?
             if [ ! -z "$exit_code" ]; then

CMakeLists.txt

@@ -29,10 +29,13 @@ option(CCCL_ENABLE_LIBCUDACXX "Enable the libcu++ developer build." ON)
 option(CCCL_ENABLE_CUB "Enable the CUB developer build." ON)
 option(CCCL_ENABLE_THRUST "Enable the Thrust developer build." ON)
 option(CCCL_ENABLE_TESTING "Enable CUDA C++ Core Library tests." ON)
+option(CCCL_ENABLE_EXAMPLES "Enable CUDA C++ Core Library examples." ON)
 
 include(CTest)
 enable_testing()
 
+include(cmake/CCCLUtilities.cmake)
+
 if (CCCL_ENABLE_LIBCUDACXX)
   set(LIBCUDACXX_TOPLEVEL_PROJECT ON)
 endif()
@@ -53,5 +56,9 @@ if (CCCL_ENABLE_TESTING)
   add_subdirectory(test)
 endif()
 
+if (CCCL_ENABLE_EXAMPLES)
+  add_subdirectory(examples)
+endif()
+
 # Must stay at the end of this file.
 include(cmake/CCCLHideThirdPartyOptions.cmake)