57 const sham::DeviceScheduler_ptr &sched,
58 const sham::DeviceBuffer<T> &buf1,
95 const sham::DeviceScheduler_ptr &sched,
96 const sham::DeviceBuffer<T> &buf1,
133 const sham::DeviceScheduler_ptr &sched,
134 const sham::DeviceBuffer<T> &buf1,
std::uint32_t u32
32 bit unsigned integer
namespace to control implementation behavior
void set_impl_reduction(const std::string &impl, const std::string ¶m="")
Set the implementation for reduction.
std::vector< shamalgs::impl_param > get_default_impl_list_reduction()
Get list of available reduction implementations.
shamalgs::impl_param get_current_impl_reduction()
Get the current implementation for reduction.
namespace for primitive algorithm (e.g. sort, scan, reductions, ...)
T sum(const sham::DeviceScheduler_ptr &sched, const sham::DeviceBuffer< T > &buf1, u32 start_id, u32 end_id)
Compute the sum of elements in a device buffer within a specified range.
T min(const sham::DeviceScheduler_ptr &sched, const sham::DeviceBuffer< T > &buf1, u32 start_id, u32 end_id)
Find the minimum element in a device buffer within a specified range.
T max(const sham::DeviceScheduler_ptr &sched, const sham::DeviceBuffer< T > &buf1, u32 start_id, u32 end_id)
Find the maximum element in a device buffer within a specified range.