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
116 lines
2.6 KiB
ReStructuredText
116 lines
2.6 KiB
ReStructuredText
.. _libcudacxx-extended-api-math:
|
|
|
|
Math
|
|
====
|
|
|
|
.. toctree::
|
|
:hidden:
|
|
:maxdepth: 1
|
|
|
|
math/ceil_div
|
|
math/round_up
|
|
math/round_down
|
|
math/ilog
|
|
math/ipow
|
|
math/pow2
|
|
math/isqrt
|
|
math/neg
|
|
math/uabs
|
|
math/fast_mod_div
|
|
math/mul_hi
|
|
math/sincos
|
|
|
|
.. list-table::
|
|
:widths: 25 45 30 30
|
|
:header-rows: 1
|
|
|
|
* - **Header**
|
|
- **Content**
|
|
- **CCCL Availability**
|
|
- **CUDA Toolkit Availability**
|
|
|
|
* - :ref:`ceil_div <libcudacxx-extended-api-math-ceil-div>`
|
|
- Ceiling division
|
|
- CCCL 2.7.0
|
|
- CUDA 12.8
|
|
|
|
* - :ref:`round_up <libcudacxx-extended-api-math-round-up>`
|
|
- Round up to the next multiple
|
|
- CCCL 3.0.0
|
|
- CUDA 13.0
|
|
|
|
* - :ref:`round_down <libcudacxx-extended-api-math-round-down>`
|
|
- Round down to the previous multiple
|
|
- CCCL 3.0.0
|
|
- CUDA 13.0
|
|
|
|
* - :ref:`ilog2 <libcudacxx-extended-api-math-ilog>`
|
|
- Integer logarithm to the base 2
|
|
- CCCL 3.0.0
|
|
- CUDA 13.0
|
|
|
|
* - :ref:`ceil_ilog2 <libcudacxx-extended-api-math-ilog>`
|
|
- Integer logarithm to the base 2, rounded up
|
|
- CCCL 3.1.0
|
|
- CUDA 13.1
|
|
|
|
* - :ref:`ilog10 <libcudacxx-extended-api-math-ilog>`
|
|
- Integer logarithm to the base 10
|
|
- CCCL 3.0.0
|
|
- CUDA 13.0
|
|
|
|
* - :ref:`ceil_ilog10 <libcudacxx-extended-api-math-ilog>`
|
|
- Integer logarithm to the base 10, rounded up
|
|
- CCCL 3.5.0
|
|
- CUDA 13.5
|
|
|
|
* - :ref:`ipow <libcudacxx-extended-api-math-ipow>`
|
|
- Integer power
|
|
- CCCL 3.1.0
|
|
- CUDA 13.1
|
|
|
|
* - :ref:`is_power_of_two <libcudacxx-extended-api-math-pow2>`
|
|
- If the value is a power of two
|
|
- CCCL 3.1.0
|
|
- CUDA 13.1
|
|
|
|
* - :ref:`isqrt <libcudacxx-extended-api-math-isqrt>`
|
|
- Integer square root
|
|
- CCCL 3.1.0
|
|
- CUDA 13.1
|
|
|
|
* - :ref:`neg <libcudacxx-extended-api-math-neg>`
|
|
- Integer negation
|
|
- CCCL 3.1.0
|
|
- CUDA 13.1
|
|
|
|
* - :ref:`next_power_of_two <libcudacxx-extended-api-math-pow2>`
|
|
- Next power of two
|
|
- CCCL 3.1.0
|
|
- CUDA 13.1
|
|
|
|
* - :ref:`prev_power_of_two <libcudacxx-extended-api-math-pow2>`
|
|
- Previous power of two
|
|
- CCCL 3.1.0
|
|
- CUDA 13.1
|
|
|
|
* - :ref:`uabs <libcudacxx-extended-api-math-uabs>`
|
|
- Unsigned absolute value
|
|
- CCCL 3.1.0
|
|
- CUDA 13.1
|
|
|
|
* - :ref:`fast_mod_div <libcudacxx-extended-api-math-fast-mod-div>`
|
|
- Fast Modulo/Division
|
|
- CCCL 3.1.0
|
|
- CUDA 13.1
|
|
|
|
* - :ref:`mul_hi <libcudacxx-extended-api-math-mul-hi>`
|
|
- Most significant half of the product
|
|
- CCCL 3.2.0
|
|
- CUDA 13.2
|
|
|
|
* - :ref:`sincos <libcudacxx-extended-api-math-sincos>`
|
|
- Computes sine and cosine of a value at the same time.
|
|
- CCCL 3.3.0
|
|
- CUDA 13.3
|