![]() |
Shamrock 2025.10.0
Astrophysical Code
|
Sort by keys algorithms. More...
#include "shambackends/DeviceBuffer.hpp"#include "shambackends/DeviceQueue.hpp"#include "shambackends/DeviceScheduler.hpp"#include <string>#include <vector>Go to the source code of this file.
Namespaces | |
| namespace | shamalgs |
| namespace to contain everything implemented by shamalgs | |
| namespace | shamalgs::primitives |
| namespace for primitive algorithm (e.g. sort, scan, reductions, ...) | |
| namespace | shamalgs::primitives::impl |
| namespace to control implementation behavior | |
Functions | |
| template<class Tkey, class Tval> | |
| void | shamalgs::primitives::sort_by_key_pow2_len (sycl::queue &q, sycl::buffer< Tkey > &buf_key, sycl::buffer< Tval > &buf_values, u32 len) |
| Sort key-value pairs using sycl::buffers (power-of-2 optimized). | |
| template<class Tkey, class Tval> | |
| void | shamalgs::primitives::sort_by_key_pow2_len (const sham::DeviceScheduler_ptr &sched, sham::DeviceBuffer< Tkey > &buf_key, sham::DeviceBuffer< Tval > &buf_values, u32 len) |
| Sort key-value pairs using USM buffers (power-of-2 optimized). | |
| std::vector< std::string > | shamalgs::primitives::impl::get_default_impl_list_sort_by_key_pow2_len () |
| Get list of available sort by key pow2 len implementations, as config json strings. | |
| std::string | shamalgs::primitives::impl::get_current_impl_sort_by_key_pow2_len () |
| Get the current implementation for sort by key pow2 len, as a config json string. | |
| bool | shamalgs::primitives::impl::is_impl_set_sort_by_key_pow2_len () |
| Check if an implementation has been selected for sort by key pow2 len. | |
| void | shamalgs::primitives::impl::set_impl_sort_by_key_pow2_len (const std::string &impl) |
| Set the implementation for sort by key pow2 len, from a config json string. | |
| void | shamalgs::primitives::impl::autoselect_impl_sort_by_key_pow2_len (const sham::DeviceScheduler_ptr &dev_sched) |
| Select the default implementation for sort by key pow2 len. | |
Sort by keys algorithms.
This header provides parallel sorting algorithms that sort key-value pairs based on the key values. The algorithms are optimized for GPU execution using sycl::buffers or USM.
sort_by_key_pow2_len only supports buffer lengths that are a power of 2 (callers must round up beforehand, e.g. with shambase::roundup_pow2).
The USM (sham::DeviceBuffer) overload selects its implementation through the generic implementation selector mechanism (see ImplVariant.hpp), via the impl sub-namespace below.
Definition in file sort_by_key_pow2_len.hpp.