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
38 lines
1002 B
ReStructuredText
38 lines
1002 B
ReStructuredText
.. _libcudacxx-ptx-instructions-mapa:
|
|
|
|
mapa
|
|
====
|
|
|
|
- PTX ISA:
|
|
`mapa <https://docs.nvidia.com/cuda/parallel-thread-execution/index.html#data-movement-and-conversion-instructions-mapa>`__
|
|
|
|
This instruction can `currently not be
|
|
implemented <https://github.com/NVIDIA/cccl/issues/1414>`__ by libcu++.
|
|
The instruction can be accessed through the cooperative groups
|
|
`cluster_group <https://docs.nvidia.com/cuda/cuda-c-programming-guide/index.html#cluster-group>`__
|
|
API:
|
|
|
|
Usage:
|
|
------
|
|
|
|
.. code:: cuda
|
|
|
|
#include <cooperative_groups.h>
|
|
|
|
__cluster_dims__(2)
|
|
__global__ void kernel() {
|
|
__shared__ int x;
|
|
x = 1;
|
|
namespace cg = cooperative_groups;
|
|
cg::cluster_group cluster = cg::this_cluster();
|
|
|
|
cluster.sync();
|
|
|
|
// Get address of remote shared memory value:
|
|
unsigned int other_block_rank = cluster.block_rank() ^ 1;
|
|
int * remote_x = cluster.map_shared_rank(&bar, other_block_rank);
|
|
|
|
// Write to remote value:
|
|
*remote_x = 2;
|
|
}
|