Files
project_6/cccl_upstream/thrust/thrust/functional.h
EngineX CI 56fd68e7dd [INFRA] Import NVIDIA/CCCL upstream as optimization reference library
CCCL (CUDA C++ Core Libraries) provides:
- CUB: device/block/warp-level GPU primitives (reduce, scan, sort, topk)
- Thrust: high-level parallel algorithms (transform_reduce, sort, scan)
- libcudacxx: CUDA C++ standard library (atomics, barriers, memory)
- cudax: experimental features (memory resources, allocators)
- Tuning policies: per-SM hardware-specific algorithm parameters

Competition optimization vectors mapped to CCCL:
- Output TPS (83% weight): warp_reduce, block_reduce, device_topk
- Input TPS (14% weight): device_scan, block_load, prefetch
- Cache TPS (3% weight): prefix caching strategy patterns
- Memory (0.9 util): pooled/cached/buddy allocators

Source: https://github.com/NVIDIA/cccl (shallow clone, HEAD only)
License: Apache-2.0
2026-07-30 09:35:51 +00:00

471 lines
14 KiB
C++

// SPDX-FileCopyrightText: Copyright (c) 2008-2018, NVIDIA Corporation. All rights reserved.
// SPDX-License-Identifier: Apache-2.0
/*! \file functional.h
* \brief Function objects and tools for manipulating them
*/
#pragma once
#include <thrust/detail/config.h>
#if defined(_CCCL_IMPLICIT_SYSTEM_HEADER_GCC)
# pragma GCC system_header
#elif defined(_CCCL_IMPLICIT_SYSTEM_HEADER_CLANG)
# pragma clang system_header
#elif defined(_CCCL_IMPLICIT_SYSTEM_HEADER_MSVC)
# pragma system_header
#endif // no system header
#include <thrust/detail/functional/actor.h>
#include <cuda/__functional/maximum.h>
#include <cuda/__functional/minimum.h>
#include <cuda/std/__functional/not_fn.h>
#include <cuda/std/__functional/operations.h>
#include <cuda/std/__type_traits/decay.h>
#include <cuda/std/__type_traits/is_constructible.h>
#include <cuda/std/__type_traits/is_move_constructible.h>
#include <cuda/std/__utility/forward.h>
THRUST_NAMESPACE_BEGIN
/*! \addtogroup predefined_function_objects Predefined Function Objects
* \ingroup function_objects
*/
/*! \addtogroup arithmetic_operations Arithmetic Operations
* \ingroup predefined_function_objects
* \{
*
* \verbatim embed:rst:leading-asterisk
* .. versionadded:: 2.2.0
* \endverbatim
*/
template <class T = void>
using divides CCCL_DEPRECATED_BECAUSE("Use cuda::std::divides instead") = ::cuda::std::divides<T>;
template <class T = void>
using minus CCCL_DEPRECATED_BECAUSE("Use cuda::std::minus instead") = ::cuda::std::minus<T>;
template <class T = void>
using modulus CCCL_DEPRECATED_BECAUSE("Use cuda::std::modulus instead") = ::cuda::std::modulus<T>;
template <class T = void>
using multiplies CCCL_DEPRECATED_BECAUSE("Use cuda::std::multiplies instead") = ::cuda::std::multiplies<T>;
template <class T = void>
using negate CCCL_DEPRECATED_BECAUSE("Use cuda::std::negate instead") = ::cuda::std::negate<T>;
template <class T = void>
using plus CCCL_DEPRECATED_BECAUSE("Use cuda::std::plus instead") = ::cuda::std::plus<T>;
/*! \p square is a function object. Specifically, it is an Adaptable Unary Function.
* If \c f is an object of class <tt>square<T></tt>, and \c x is an object
* of class \c T, then <tt>f(x)</tt> returns <tt>x*x</tt>.
*
* \tparam T is a model of <a href="https://en.cppreference.com/w/cpp/named_req/CopyAssignable">Assignable</a>,
* and if \c x is an object of type \p T, then <tt>x*x</tt> must be defined and must have a return type that is
* convertible to \c T.
*
* The following code snippet demonstrates how to use <tt>square</tt> to square
* the elements of a device_vector of \c floats.
*
* \code
* #include <thrust/device_vector.h>
* #include <thrust/functional.h>
* #include <thrust/sequence.h>
* #include <thrust/transform.h>
* ...
* const int N = 1000;
* thrust::device_vector<float> V1(N);
* thrust::device_vector<float> V2(N);
*
* thrust::sequence(V1.begin(), V1.end(), 1);
*
* thrust::transform(V1.begin(), V1.end(), V2.begin(),
* thrust::square<float>());
* // V2 is now {1, 4, 9, ..., 1000000}
* \endcode
*
* \verbatim embed:rst:leading-asterisk
* .. versionadded:: 2.2.0
* \endverbatim
*/
template <typename T = void>
struct square
{
_CCCL_EXEC_CHECK_DISABLE
_CCCL_HOST_DEVICE constexpr T operator()(const T& x) const
{
return x * x;
}
};
template <>
struct square<void>
{
using is_transparent = void;
_CCCL_EXEC_CHECK_DISABLE
template <typename T>
_CCCL_HOST_DEVICE constexpr T operator()(const T& x) const noexcept(noexcept(x * x))
{
return x * x;
}
};
/*! \}
*/
/*! \addtogroup comparison_operations Comparison Operations
* \ingroup predefined_function_objects
* \{
*
* \verbatim embed:rst:leading-asterisk
* .. versionadded:: 2.2.0
* \endverbatim
*/
//! deprecated [since 3.1]
template <class T = void>
using equal_to CCCL_DEPRECATED_BECAUSE("Use cuda::std::equal_to instead") = ::cuda::std::equal_to<T>;
//! deprecated [since 3.1]
template <class T = void>
using greater CCCL_DEPRECATED_BECAUSE("Use cuda::std::greater instead") = ::cuda::std::greater<T>;
//! deprecated [since 3.1]
template <class T = void>
using greater_equal CCCL_DEPRECATED_BECAUSE("Use cuda::std::greater_equal instead") = ::cuda::std::greater_equal<T>;
//! deprecated [since 3.1]
template <class T = void>
using less CCCL_DEPRECATED_BECAUSE("Use cuda::std::less instead") = ::cuda::std::less<T>;
//! deprecated [since 3.1]
template <class T = void>
using less_equal CCCL_DEPRECATED_BECAUSE("Use cuda::std::less_equal instead") = ::cuda::std::less_equal<T>;
//! deprecated [since 3.1]
template <class T = void>
using not_equal_to CCCL_DEPRECATED_BECAUSE("Use cuda::std::not_equal_to instead") = ::cuda::std::not_equal_to<T>;
/*! \}
*/
/*! \addtogroup logical_operations Logical Operations
* \ingroup predefined_function_objects
* \{
*
* \verbatim embed:rst:leading-asterisk
* .. versionadded:: 2.2.0
* \endverbatim
*/
//! deprecated [since 3.1]
template <class T = void>
using logical_and CCCL_DEPRECATED_BECAUSE("Use cuda::std::logical_and instead") = ::cuda::std::logical_and<T>;
//! deprecated [since 3.1]
template <class T = void>
using logical_not CCCL_DEPRECATED_BECAUSE("Use cuda::std::logical_not instead") = ::cuda::std::logical_not<T>;
//! deprecated [since 3.1]
template <class T = void>
using logical_or CCCL_DEPRECATED_BECAUSE("Use cuda::std::logical_or instead") = ::cuda::std::logical_or<T>;
/*! \}
*/
/*! \addtogroup bitwise_operations Bitwise Operations
* \ingroup predefined_function_objects
* \{
*
* \verbatim embed:rst:leading-asterisk
* .. versionadded:: 2.2.0
* \endverbatim
*/
//! deprecated [since 3.1]
template <class T = void>
using bit_and CCCL_DEPRECATED_BECAUSE("Use cuda::std::bit_and instead") = ::cuda::std::bit_and<T>;
//! deprecated [since 3.1]
template <class T = void>
using bit_or CCCL_DEPRECATED_BECAUSE("Use cuda::std::bit_or instead") = ::cuda::std::bit_or<T>;
//! deprecated [since 3.1]
template <class T = void>
using bit_xor CCCL_DEPRECATED_BECAUSE("Use cuda::std::bit_xor instead") = ::cuda::std::bit_xor<T>;
/*! \}
*/
/*! \addtogroup generalized_identity_operations Generalized Identity Operations
* \ingroup predefined_function_objects
* \{
*
* \verbatim embed:rst:leading-asterisk
* .. versionadded:: 2.2.0
* \endverbatim
*/
//! deprecated [since 3.1]
template <class T = void>
using maximum CCCL_DEPRECATED_BECAUSE("Use cuda::maximum instead") = ::cuda::maximum<T>;
//! deprecated [since 3.1]
template <class T = void>
using minimum CCCL_DEPRECATED_BECAUSE("Use cuda::minimum instead") = ::cuda::minimum<T>;
/*! \p project1st is a function object that takes two arguments and returns
* its first argument; the second argument is unused. It is essentially a
* generalization of identity to the case of a Binary Function.
*
* \code
* #include <thrust/functional.h>
* #include <assert.h>
* ...
* int x = 137;
* int y = -137;
* thrust::project1st<int> pj1;
* assert(x == pj1(x,y));
* \endcode
*
* \see identity
* \see project2nd
*
* \verbatim embed:rst:leading-asterisk
* .. versionadded:: 2.2.0
* \endverbatim
*/
template <typename T1 = void, typename T2 = void>
struct project1st
{
/*! Function call operator. The return value is <tt>lhs</tt>.
*
* \verbatim embed:rst:leading-asterisk
* .. versionadded:: 2.2.0
* \endverbatim
*/
_CCCL_HOST_DEVICE constexpr const T1& operator()(const T1& lhs _CCCL_LIFETIMEBOUND, const T2& /*rhs*/) const
{
return lhs;
}
};
template <>
struct project1st<void, void>
{
using is_transparent = void;
_CCCL_EXEC_CHECK_DISABLE
template <typename T1, typename T2>
_CCCL_HOST_DEVICE constexpr auto operator()(T1&& t1, T2&&) const noexcept(noexcept(THRUST_FWD(t1)))
-> decltype(THRUST_FWD(t1))
{
return THRUST_FWD(t1);
}
};
/*! \p project2nd is a function object that takes two arguments and returns
* its second argument; the first argument is unused. It is essentially a
* generalization of identity to the case of a Binary Function.
*
* \code
* #include <thrust/functional.h>
* #include <assert.h>
* ...
* int x = 137;
* int y = -137;
* thrust::project2nd<int> pj2;
* assert(y == pj2(x,y));
* \endcode
*
* \see identity
* \see project1st
*
* \verbatim embed:rst:leading-asterisk
* .. versionadded:: 2.2.0
* \endverbatim
*/
template <typename T1 = void, typename T2 = void>
struct project2nd
{
/*! Function call operator. The return value is <tt>rhs</tt>.
*
* \verbatim embed:rst:leading-asterisk
* .. versionadded:: 2.2.0
* \endverbatim
*/
_CCCL_HOST_DEVICE constexpr const T2& operator()(const T1& /*lhs*/, const T2& rhs _CCCL_LIFETIMEBOUND) const
{
return rhs;
}
}; // end project2nd
template <>
struct project2nd<void, void>
{
using is_transparent = void;
_CCCL_EXEC_CHECK_DISABLE
template <typename T1, typename T2>
_CCCL_HOST_DEVICE constexpr auto operator()(T1&&, T2&& t2) const noexcept(noexcept(THRUST_FWD(t2)))
-> decltype(THRUST_FWD(t2))
{
return THRUST_FWD(t2);
}
};
/*! \}
*/
// odds and ends
/*! \addtogroup function_object_adaptors
* \{
*/
//! deprecated [since 3.1]
#ifdef _CCCL_DOXYGEN_INVOKED
using ::cuda::std::not_fn;
#else // ^^^ _CCCL_DOXYGEN_INVOKED ^^^ / vvv !_CCCL_DOXYGEN_INVOKED vvv
_CCCL_TEMPLATE(class _Fn)
_CCCL_REQUIRES(::cuda::std::is_constructible_v<::cuda::std::decay_t<_Fn>, _Fn>
_CCCL_AND ::cuda::std::is_move_constructible_v<::cuda::std::decay_t<_Fn>>)
[[nodiscard]] CCCL_DEPRECATED_BECAUSE("Use cuda::std::not_fn instead") _CCCL_HOST_DEVICE_API constexpr auto
not_fn(_Fn&& __f)
{
return ::cuda::std::not_fn(::cuda::std::forward<_Fn>(__f));
}
#endif // !_CCCL_DOXYGEN_INVOKED
/*! \}
*/
/*! \addtogroup placeholder_objects Placeholder Objects
* \ingroup function_objects
* \{
*/
/*! \namespace thrust::placeholders
* \brief Facilities for constructing simple functions inline.
*
* Objects in the \p thrust::placeholders namespace may be used to create simple arithmetic functions inline
* in an algorithm invocation. Combining placeholders such as \p _1 and \p _2 with arithmetic operations such as \c +
* creates an unnamed function object which applies the operation to their arguments.
*
* The type of placeholder objects is implementation-defined.
*
* The following code snippet demonstrates how to use the placeholders \p _1 and \p _2 with \p thrust::transform
* to implement the SAXPY computation:
*
* \code
* #include <thrust/device_vector.h>
* #include <thrust/transform.h>
* #include <thrust/functional.h>
*
* int main()
* {
* thrust::device_vector<float> x(4), y(4);
* x[0] = 1;
* x[1] = 2;
* x[2] = 3;
* x[3] = 4;
*
* y[0] = 1;
* y[1] = 1;
* y[2] = 1;
* y[3] = 1;
*
* float a = 2.0f;
*
* using namespace thrust::placeholders;
*
* thrust::transform(x.begin(), x.end(), y.begin(), y.begin(),
* a * _1 + _2
* );
*
* // y is now {3, 5, 7, 9}
* }
* \endcode
*
* \verbatim embed:rst:leading-asterisk
* .. versionadded:: 2.2.0
* \endverbatim
*/
namespace placeholders
{
/*! \p thrust::placeholders::_1 is the placeholder for the first function parameter.
*
* \verbatim embed:rst:leading-asterisk
* .. versionadded:: 2.2.0
* \endverbatim
*/
_CCCL_GLOBAL_CONSTANT thrust::detail::functional::placeholder<0>::type _1;
/*! \p thrust::placeholders::_2 is the placeholder for the second function parameter.
*
* \verbatim embed:rst:leading-asterisk
* .. versionadded:: 2.2.0
* \endverbatim
*/
_CCCL_GLOBAL_CONSTANT thrust::detail::functional::placeholder<1>::type _2;
/*! \p thrust::placeholders::_3 is the placeholder for the third function parameter.
*
* \verbatim embed:rst:leading-asterisk
* .. versionadded:: 2.2.0
* \endverbatim
*/
_CCCL_GLOBAL_CONSTANT thrust::detail::functional::placeholder<2>::type _3;
/*! \p thrust::placeholders::_4 is the placeholder for the fourth function parameter.
*
* \verbatim embed:rst:leading-asterisk
* .. versionadded:: 2.2.0
* \endverbatim
*/
_CCCL_GLOBAL_CONSTANT thrust::detail::functional::placeholder<3>::type _4;
/*! \p thrust::placeholders::_5 is the placeholder for the fifth function parameter.
*
* \verbatim embed:rst:leading-asterisk
* .. versionadded:: 2.2.0
* \endverbatim
*/
_CCCL_GLOBAL_CONSTANT thrust::detail::functional::placeholder<4>::type _5;
/*! \p thrust::placeholders::_6 is the placeholder for the sixth function parameter.
*
* \verbatim embed:rst:leading-asterisk
* .. versionadded:: 2.2.0
* \endverbatim
*/
_CCCL_GLOBAL_CONSTANT thrust::detail::functional::placeholder<5>::type _6;
/*! \p thrust::placeholders::_7 is the placeholder for the seventh function parameter.
*
* \verbatim embed:rst:leading-asterisk
* .. versionadded:: 2.2.0
* \endverbatim
*/
_CCCL_GLOBAL_CONSTANT thrust::detail::functional::placeholder<6>::type _7;
/*! \p thrust::placeholders::_8 is the placeholder for the eighth function parameter.
*
* \verbatim embed:rst:leading-asterisk
* .. versionadded:: 2.2.0
* \endverbatim
*/
_CCCL_GLOBAL_CONSTANT thrust::detail::functional::placeholder<7>::type _8;
/*! \p thrust::placeholders::_9 is the placeholder for the ninth function parameter.
*
* \verbatim embed:rst:leading-asterisk
* .. versionadded:: 2.2.0
* \endverbatim
*/
_CCCL_GLOBAL_CONSTANT thrust::detail::functional::placeholder<8>::type _9;
/*! \p thrust::placeholders::_10 is the placeholder for the tenth function parameter.
*
* \verbatim embed:rst:leading-asterisk
* .. versionadded:: 2.2.0
* \endverbatim
*/
_CCCL_GLOBAL_CONSTANT thrust::detail::functional::placeholder<9>::type _10;
} // namespace placeholders
/*! \} // placeholder_objects
*/
THRUST_NAMESPACE_END
#include <thrust/detail/functional/operators.h>