Remove the existing cuda.coop implementation

.pre-commit-config.yaml

@@ -79,21 +79,7 @@ repos:
     rev: v0.15.20
     hooks:
     - id: ruff  # linter
-      exclude: ^python/cuda_coop/
     - id: ruff-format  # formatter
-      exclude: ^python/cuda_coop/
-
-  # Use the newer Ruff rules for `cuda.coop`.
-  - repo: https://github.com/astral-sh/ruff-pre-commit
-    rev: v0.16.9
-    hooks:
-    - id: ruff-check
-      name: ruff check (cuda.coop)
-      files: ^python/cuda_coop/
-    - id: ruff-format
-      name: ruff format (cuda.coop)
-      files: ^python/cuda_coop/
-      types_or: [python, pyi, jupyter]
 
   # CMake formatting
   - repo: https://github.com/BlankSpruce/gersemi

.vscode/launch.json

@@ -1,39 +1,5 @@
 {
   "configurations": [
-    {
-      "console": "integratedTerminal",
-      "cwd": "${workspaceFolder}",
-      "env": {
-        "CUDA_COOP_CCCL_ROOT": "${workspaceFolder}",
-        "CUDA_COOP_DISABLE_AUTO_DSL_REGISTRATION": "0",
-        "CUDA_COOP_ENABLE_CACHE": "0",
-        "CUDA_COOP_SOURCE_DUMP_DIR": "${workspaceFolder}/build/coop-debug-sources",
-        "CUDA_VISIBLE_DEVICES": "0",
-        "PYTHONPATH": "${workspaceFolder}/python/cuda_coop"
-      },
-      "justMyCode": false,
-      "name": "cuda.coop: Debugger walkthrough",
-      "program": "${workspaceFolder}/docs/python/coop/debugger_walkthrough.py",
-      "request": "launch",
-      "type": "debugpy"
-    },
-    {
-      "console": "integratedTerminal",
-      "cwd": "${workspaceFolder}",
-      "env": {
-        "CUDA_COOP_CCCL_ROOT": "${workspaceFolder}",
-        "CUDA_COOP_DISABLE_AUTO_DSL_REGISTRATION": "0",
-        "CUDA_COOP_ENABLE_CACHE": "0",
-        "CUDA_COOP_SOURCE_DUMP_DIR": "${workspaceFolder}/build/coop-debug-sources",
-        "CUDA_VISIBLE_DEVICES": "0",
-        "PYTHONPATH": "${workspaceFolder}/python/cuda_coop"
-      },
-      "justMyCode": false,
-      "name": "cuda.coop: Debug active Python file",
-      "program": "${file}",
-      "request": "launch",
-      "type": "debugpy"
-    },
     {
       "cwd": "${workspaceFolder}",
       "initCommands": [

AGENTS.md

@@ -2,10 +2,6 @@
 
 This document provides guidelines for building, testing, and contributing to the CCCL repository. It is primarily written for agentic AIs, but the information is also useful for CCCL developers.
 
-For `cuda.coop` work, including `python/cuda_coop/**` and
-`docs/python/coop*`, read [python/cuda_coop/AGENTS.md](python/cuda_coop/AGENTS.md)
-and apply its API, behavior, and example conventions.
-
 ---
 
 ## Overview

ci/matrix.yaml

@@ -214,12 +214,6 @@ workflows:
     - {jobs: ['test_py_compute_minimal'], project: 'python_tsan', ctk: '13.X', py_version: '3.14t', gpu: 'l4', cxx: 'gcc13'}
     # cuda.cccl.headers: CPU-only (gpu:false), arch-insensitive
     - {jobs: ['test_headers'], project: 'python', ctk: ['12.X', '13.X'], py_version: '3.14', cxx: ['gcc13', 'msvc2022'], py_ctk_mode: ['pinned', 'latest', 'sysctk']}
-    # cuda.coop: installed-wheel host contracts, GPU-free compilation, one
-    # representative runtime lane, and Windows universal-wheel endpoints.
-    - {jobs: ['test_py_coop_host'], project: 'cuda_coop_python', ctk: '13.X', py_version: ['3.10', '3.14'], cxx: 'gcc13'}
-    - {jobs: ['test_py_coop_numba_mlir_compile'], project: 'cuda_coop_python', ctk: '13.X', py_version: '3.14', cxx: 'gcc13'}
-    - {jobs: ['test_py_coop_numba_mlir_runtime'], project: 'cuda_coop_python', ctk: '13.X', py_version: '3.14', gpu: 'l4', cxx: 'gcc13'}
-    - {jobs: ['build'], project: 'cuda_coop_python', ctk: '13.X', py_version: ['3.10', '3.14'], cxx: 'msvc2022'}
     # CCCL packaging:
     - {jobs: ['test'], project: 'packaging', ctk: '12.0', cxx: ['gcc10', 'clang14'], gpu: 't4', args: '-min-cmake'}
     - {jobs: ['test'], project: 'packaging', ctk: '12.X', cxx: ['gcc10', 'clang14'], gpu: 't4'}
@@ -286,10 +280,6 @@ workflows:
     - {project: 'python', jobs: ['test'], ctk: '13.X', py_version: '3.14', gpu: 'l4', cxx: ['gcc13', 'msvc2022']}
     - {project: 'python', jobs: ['test_headers'], ctk: '13.X', py_version: '3.14', cxx: ['gcc13', 'msvc2022']}
     - {project: 'python', jobs: ['test_py_stf'], ctk: '13.X', py_version: '3.14', gpu: 'l4', cxx: 'gcc13'}
-    # Test the installed `cuda-coop` wheel when its dependencies change.
-    - {project: 'cuda_coop_python', jobs: ['test_py_coop_host'], ctk: '13.X', py_version: '3.14', cxx: 'gcc13'}
-    - {project: 'cuda_coop_python', jobs: ['test_py_coop_numba_mlir_compile'], ctk: '13.X', py_version: '3.14', cxx: 'gcc13'}
-    - {project: 'cuda_coop_python', jobs: ['test_py_coop_numba_mlir_runtime'], ctk: '13.X', py_version: '3.14', gpu: 'l4', cxx: 'gcc13'}
     # Packaging / install
     - {project: 'packaging', jobs: ['test'], ctk: '13.X', cxx: ['gcc', 'clang'], gpu: 't4', sm: 'gpu'}
     - {project: 'packaging', jobs: ['test'], args: '-min-cmake', gpu: 't4', sm: 'gpu'}
@@ -437,13 +427,6 @@ workflows:
     - {jobs: ['test_py_compute_minimal'], project: 'python', ctk: ['12.0', '12.X', '13.0', '13.X'], py_version: '3.14t', gpu: 'l4', cxx: ['gcc13', 'msvc2022'], py_ctk_mode: 'sysctk'}
     # cuda.cccl.headers (CPU-only): all CTK x source x OS, py endpoints 3.10 + 3.14
     - {jobs: ['test_headers'], project: 'python', ctk: ['12.0', '12.X', '13.0', '13.X'], py_version: ['3.10', '3.14'], cxx: ['gcc13', 'msvc2022'], py_ctk_mode: ['pinned', 'latest', 'sysctk']}
-    # cuda.coop: run the older supported CUDA major in `nightly`. The CUDA 13
-    # compiler/runtime counterparts run in `pull_request`, while the pure wheel
-    # is checked at both Python endpoints on Linux and Windows in both.
-    - {jobs: ['test_py_coop_host'], project: 'cuda_coop_python', ctk: '13.X', py_version: ['3.10', '3.14'], cxx: 'gcc13'}
-    - {jobs: ['test_py_coop_numba_mlir_compile'], project: 'cuda_coop_python', ctk: '12.X', py_version: '3.14', cxx: 'gcc13'}
-    - {jobs: ['test_py_coop_numba_mlir_runtime'], project: 'cuda_coop_python', ctk: '12.X', py_version: '3.14', gpu: 'l4', cxx: 'gcc13'}
-    - {jobs: ['build'], project: 'cuda_coop_python', ctk: '13.X', py_version: ['3.10', '3.14'], cxx: 'msvc2022'}
     # CCCL packaging:
     - {jobs: ['test'], project: 'packaging', ctk: '12.0', cxx: ['gcc10', 'clang14'], gpu: 't4', args: '-min-cmake'}
     - {jobs: ['test'], project: 'packaging', ctk: '12.X', cxx: ['gcc10', 'clang14'], gpu: 'rtx2080'}
@@ -607,8 +590,7 @@ workflows:
     # deliberately absent (like the TSan lanes below): the experimental
     # cuda-stf wheel must not be published, and the combined STF producer's
     # cuda_cccl wheel would collide with the release one in dist/. STF
-    # coverage lives in the pull_request/nightly/weekly workflows. cuda.coop
-    # is also absent until cuda-coop has a registered release target.
+    # coverage lives in the pull_request/nightly/weekly workflows.
     - {jobs: ['test'], project: 'python', ctk: ['12.0', '12.X', '13.0', '13.X'], py_version: ['3.10', '3.11', '3.12', '3.13', '3.14', '3.14t'], gpu: 'l4', cxx: 'gcc13'}
     - {jobs: ['test'], project: 'python', ctk: ['12.X', '13.0', '13.X'], py_version: ['3.10', '3.11', '3.12', '3.13', '3.14', '3.14t'], gpu: 'l4', cxx: 'msvc2022'}
     # Windows + CTK 12.0: compute only, no examples (NVBug 6706273; see the nightly comment).
@@ -856,9 +838,6 @@ jobs:
   test_py_compute_minimal: { name: "Test cuda.compute minimal", gpu: true, needs: 'build_py_wheel', force_producer_ctk: "pybuild", invoke: { prefix: 'test_cuda_compute_minimal'} }
   test_py_examples: { name: "Test cuda.cccl.examples",     gpu: true,  needs: 'build_py_wheel', force_producer_ctk: "pybuild", invoke: { prefix: 'test_cuda_cccl_examples'} }
   test_py_stf:      { name: "Test cuda.stf._experimental", gpu: true,  needs: 'build_py_stf_wheel', force_producer_ctk: "pybuild", invoke: { prefix: 'test_cuda_stf'} }
-  test_py_coop_host: { name: "Test cuda.coop (host)", gpu: false, needs: 'build', force_producer_ctk: "pybuild", invoke: { prefix: 'test', args: '-stage contracts'} }
-  test_py_coop_numba_mlir_compile: { name: "Compile cuda.coop Numba-CUDA-MLIR (GPU hidden)", gpu: false, needs: 'build', force_producer_ctk: "pybuild", invoke: { prefix: 'test', args: '-stage numba-mlir-compile'} }
-  test_py_coop_numba_mlir_runtime: { name: "Test cuda.coop Numba-CUDA-MLIR (L4)", gpu: true, needs: 'build', force_producer_ctk: "pybuild", invoke: { prefix: 'test', args: '-stage numba-mlir-runtime'} }
 
   # Run jobs for 'target' project (ci/util/build_and_test_targets.sh):
   run_cpu: { gpu: false }
@@ -942,8 +921,6 @@ projects:
     job_map:
       build: ['build_py_wheel']
       test:  ['test_py_compute_minimal']
-  cuda_coop_python:
-    name: "Python (cuda.coop)"
   cccl_c_parallel:
     name: 'CCCL C Parallel'
     stds: [20]

ci/project_files_and_dependencies.yaml

@@ -54,7 +54,6 @@ projects:
     lite_dependencies: [c2h]
     full_dependencies: [libcudacxx_public]
     include_regexes: ["libcudacxx/"]
-    exclude_regexes: ["^libcudacxx/LICENSE\\.TXT$"]
     exclude_project_files: [libcudacxx_public]
 
   cub_public:
@@ -69,7 +68,6 @@ projects:
     lite_dependencies: [c2h, nvbench_helper]
     full_dependencies: [cub_public]
     include_regexes: ["cub/"]
-    exclude_regexes: ["^cub/LICENSE\\.TXT$"]
     exclude_project_files: [cub_public]
 
   thrust_public:
@@ -84,7 +82,6 @@ projects:
     lite_dependencies: [nvbench_helper]
     full_dependencies: [thrust_public]
     include_regexes: ["thrust/"]
-    exclude_regexes: ["^thrust/LICENSE$"]
     exclude_project_files: [thrust_public]
 
   cudax_public:
@@ -99,7 +96,6 @@ projects:
     lite_dependencies: [c2h, nvbench_helper]
     full_dependencies: [cudax_public]
     include_regexes: ["cudax/"]
-    exclude_regexes: ["^cudax/LICENSE\\.TXT$"]
     exclude_project_files: [cudax_public]
 
   cccl_c_parallel_public:
@@ -176,29 +172,6 @@ projects:
     - "pyproject.toml"
     exclude_regexes: []
 
-  cuda_coop_python:
-    name: "cuda.coop Python"
-    matrix_project: "cuda_coop_python"
-    # The wheel carries these three public header trees at one CCCL revision.
-    # Header changes therefore exercise the focused installed-wheel lane.
-    lite_dependencies:
-      - libcudacxx_public
-      - cub_public
-      - thrust_public
-    full_dependencies: []
-    include_regexes:
-    - "python/cuda_coop/"
-    - "ci/build_cuda_coop_python\\.sh"
-    - "ci/test_cuda_coop_python\\.sh"
-    - "ci/validate_cuda_coop_wheel\\.py"
-    - "ci/windows/build_cuda_coop_python\\.ps1"
-    - "^LICENSE$"
-    - "^cub/LICENSE\\.TXT$"
-    - "^libcudacxx/LICENSE\\.TXT$"
-    - "^thrust/LICENSE$"
-    - "^pyproject\\.toml$"
-    exclude_regexes: []
-
   packaging:
     name: "CCCL Packaging"
     matrix_project: "packaging"
@@ -274,8 +247,6 @@ projects:
 
 # Files matching any of these regexes will be ignored globally.
 ignore_regexes:
-  - '^cudax/LICENSE\.TXT$'
-  - '^libcudacxx/test/LICENSE\.TXT$'
   - '.+\.md$'
   - '\.branch_notes/'
   - '^\.agent/'
@@ -334,3 +305,9 @@ ignore_regexes:
   - 'ci/rapids/'
   - '^ci/util/pre-commit/'
   - 'docs/'
+  - '^LICENSE$'
+  - '^cub/LICENSE\.TXT$'
+  - '^cudax/LICENSE\.TXT$'
+  - '^libcudacxx/LICENSE\.TXT$'
+  - '^libcudacxx/test/LICENSE\.TXT$'
+  - '^thrust/LICENSE$'

ci/test/inspect_changes/core_dirty.output

@@ -1,2 +1,2 @@
-FULL_BUILD=libcudacxx cub thrust cudax cccl_c_parallel cccl_c_parallel_v2 python_v2 python_tsan cccl_c_stf python cuda_coop_python packaging stdpar nvbench_helper nvrtcc tidy
+FULL_BUILD=libcudacxx cub thrust cudax cccl_c_parallel cccl_c_parallel_v2 python_v2 python_tsan cccl_c_stf python packaging stdpar nvbench_helper nvrtcc tidy
 LITE_BUILD=

ci/test/inspect_changes/ignored_only.dirty_files

@@ -1,2 +1,7 @@
 docs/index.rst
+LICENSE
+cub/LICENSE.TXT
 cudax/LICENSE.TXT
+libcudacxx/LICENSE.TXT
+libcudacxx/test/LICENSE.TXT
+thrust/LICENSE

ci/test/inspect_changes/libcudacxx_both.output

@@ -1,2 +1,2 @@
 FULL_BUILD=libcudacxx tidy
-LITE_BUILD=cub thrust cudax cccl_c_parallel cccl_c_parallel_v2 python_v2 python_tsan cccl_c_stf python cuda_coop_python packaging stdpar nvbench_helper
+LITE_BUILD=cub thrust cudax cccl_c_parallel cccl_c_parallel_v2 python_v2 python_tsan cccl_c_stf python packaging stdpar nvbench_helper

ci/test/inspect_changes/libcudacxx_public_only.output

@@ -1,2 +1,2 @@
 FULL_BUILD=libcudacxx tidy
-LITE_BUILD=cub thrust cudax cccl_c_parallel cccl_c_parallel_v2 python_v2 python_tsan cccl_c_stf python cuda_coop_python packaging stdpar nvbench_helper
+LITE_BUILD=cub thrust cudax cccl_c_parallel cccl_c_parallel_v2 python_v2 python_tsan cccl_c_stf python packaging stdpar nvbench_helper

ci/test/inspect_changes/libcudacxx_thrust.output

@@ -1,2 +1,2 @@
 FULL_BUILD=libcudacxx thrust tidy
-LITE_BUILD=cub cudax cccl_c_parallel cccl_c_parallel_v2 python_v2 python_tsan cccl_c_stf python cuda_coop_python packaging stdpar nvbench_helper
+LITE_BUILD=cub cudax cccl_c_parallel cccl_c_parallel_v2 python_v2 python_tsan cccl_c_stf python packaging stdpar nvbench_helper