2a7ca101d7
Sparse-checkout from NVIDIA/cccl main branch to complete cccl_upstream: Added: - python/cuda_cccl/ (226 files) — Python bindings for device-level algorithms Critical for muh toolchain: cuda.compute.reduce_into, scan, radix_sort, etc. Includes 204 .py files with full test coverage for all 27 algorithms - ci/ (163 files) — Build/test infrastructure build_cub.sh, test_cub.sh, build_and_test_targets.sh, matrix.yaml Directly maps to our [INFRA-CI] and [INFRA-BUILD] items - .agent/skills/ (7 files) — NVIDIA's own agent skills for CCCL cccl-style/SKILL.md, cccl-test/SKILL.md, sass-diff/SKILL.md - docs/ (491 files) — Official CCCL documentation CI references, CMake guides, Python compute docs, libcudacxx PTX docs - test/ (12 files) — Top-level integration tests (cuda_smoke, stdpar) - Root configs: .clang-format, .clang-tidy, CONTRIBUTING.md, pyproject.toml - CLAUDE.md symlink → AGENTS.md (NVIDIA's standard) cccl_upstream now mirrors full NVIDIA/cccl structure: Before: 42M (cub + thrust + libcudacxx + cudax + c + examples + benchmarks) After: 53M (+python +ci +docs +.agent +test +configs) This completes the CCCL base needed for: - [muh-bench] items: ci/util/build_and_test_targets.sh for targeted builds - [CCCL-verify] items: python/cuda_cccl/tests/ as reference implementations - [CCCL-test] items: ci/test_cub.sh, ci/test_thrust.sh - Agent workflow: .agent/skills/ for consistent style and test patterns
69 lines
3.5 KiB
ReStructuredText
69 lines
3.5 KiB
ReStructuredText
..
|
|
This file was automatically generated. Do not edit.
|
|
|
|
clusterlaunchcontrol.try_cancel.async.shared::cta.mbarrier::complete_tx::bytes.b128
|
|
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
|
|
.. code-block:: cuda
|
|
|
|
// clusterlaunchcontrol.try_cancel.async.shared::cta.mbarrier::complete_tx::bytes.b128 [addr], [smem_bar]; // PTX ISA 86, SM_100
|
|
template <typename = void>
|
|
__device__ static inline void clusterlaunchcontrol_try_cancel(
|
|
void* addr,
|
|
uint64_t* smem_bar);
|
|
|
|
clusterlaunchcontrol.try_cancel.async.shared::cta.mbarrier::complete_tx::bytes.multicast::cluster::all.b128
|
|
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
|
|
.. code-block:: cuda
|
|
|
|
// clusterlaunchcontrol.try_cancel.async.shared::cta.mbarrier::complete_tx::bytes.multicast::cluster::all.b128 [addr], [smem_bar]; // PTX ISA 86, SM_100a, SM_110a
|
|
template <typename = void>
|
|
__device__ static inline void clusterlaunchcontrol_try_cancel_multicast(
|
|
void* addr,
|
|
uint64_t* smem_bar);
|
|
|
|
clusterlaunchcontrol.query_cancel.is_canceled.pred.b128
|
|
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
|
|
.. code-block:: cuda
|
|
|
|
// clusterlaunchcontrol.query_cancel.is_canceled.pred.b128 pred_is_canceled, try_cancel_response; // PTX ISA 86, SM_100
|
|
template <typename B128, enable_if_t<sizeof(B128) == 16, bool> = true>
|
|
__device__ static inline bool clusterlaunchcontrol_query_cancel_is_canceled(
|
|
B128 try_cancel_response);
|
|
|
|
clusterlaunchcontrol.query_cancel.get_first_ctaid::x.b32.b128
|
|
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
|
|
.. code-block:: cuda
|
|
|
|
// clusterlaunchcontrol.query_cancel.get_first_ctaid::x.b32.b128 ret_dim, try_cancel_response; // PTX ISA 86, SM_100
|
|
template <typename B32, enable_if_t<sizeof(B32) == 4, bool> = true, typename B128, enable_if_t<sizeof(B128) == 16, bool> = true>
|
|
__device__ static inline B32 clusterlaunchcontrol_query_cancel_get_first_ctaid_x(
|
|
B128 try_cancel_response);
|
|
|
|
clusterlaunchcontrol.query_cancel.get_first_ctaid::y.b32.b128
|
|
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
|
|
.. code-block:: cuda
|
|
|
|
// clusterlaunchcontrol.query_cancel.get_first_ctaid::y.b32.b128 ret_dim, try_cancel_response; // PTX ISA 86, SM_100
|
|
template <typename B32, enable_if_t<sizeof(B32) == 4, bool> = true, typename B128, enable_if_t<sizeof(B128) == 16, bool> = true>
|
|
__device__ static inline B32 clusterlaunchcontrol_query_cancel_get_first_ctaid_y(
|
|
B128 try_cancel_response);
|
|
|
|
clusterlaunchcontrol.query_cancel.get_first_ctaid::z.b32.b128
|
|
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
|
|
.. code-block:: cuda
|
|
|
|
// clusterlaunchcontrol.query_cancel.get_first_ctaid::z.b32.b128 ret_dim, try_cancel_response; // PTX ISA 86, SM_100
|
|
template <typename B32, enable_if_t<sizeof(B32) == 4, bool> = true, typename B128, enable_if_t<sizeof(B128) == 16, bool> = true>
|
|
__device__ static inline B32 clusterlaunchcontrol_query_cancel_get_first_ctaid_z(
|
|
B128 try_cancel_response);
|
|
|
|
clusterlaunchcontrol.query_cancel.get_first_ctaid.v4.b32.b128
|
|
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
|
|
.. code-block:: cuda
|
|
|
|
// clusterlaunchcontrol.query_cancel.get_first_ctaid.v4.b32.b128 block_dim, try_cancel_response; // PTX ISA 86, SM_100
|
|
template <typename B32, enable_if_t<sizeof(B32) == 4, bool> = true, typename B128, enable_if_t<sizeof(B128) == 16, bool> = true>
|
|
__device__ static inline void clusterlaunchcontrol_query_cancel_get_first_ctaid(
|
|
B32 (&block_dim)[4],
|
|
B128 try_cancel_response);
|