58 const sham::DeviceScheduler_ptr &sched,
59 const sham::DeviceBuffer<T> &buf1,
96 const sham::DeviceScheduler_ptr &sched,
97 const sham::DeviceBuffer<T> &buf1,
134 const sham::DeviceScheduler_ptr &sched,
135 const sham::DeviceBuffer<T> &buf1,
std::uint32_t u32
32 bit unsigned integer
namespace to control implementation behavior
bool is_impl_set_reduction()
Check if an implementation has been selected for reduction.
void autoselect_impl_reduction(const sham::DeviceScheduler_ptr &dev_sched)
Select the default implementation for reduction.
std::string get_current_impl_reduction()
Get the current implementation for reduction, as a config json string.
std::vector< std::string > get_default_impl_list_reduction()
Get list of available reduction implementations, as config json strings.
void set_impl_reduction(const std::string &impl)
Set the implementation for reduction, from a config json string.
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.