/home/docs/checkouts/readthedocs.org/user_builds/advanced-micro-devices-composable-kernel/checkouts/develop/include/ck/tensor_operation/gpu/grid/gridwise_gemm_pipeline_selector.hpp File Reference

• develop
• include
• ck
• tensor_operation
• gpu
• grid

Namespaces | Enumerations | Functions

gridwise_gemm_pipeline_selector.hpp File Reference

#include <iostream>
#include <ostream>
#include "ck/tensor_operation/gpu/grid/gridwise_gemm_pipeline_v1.hpp"
#include "ck/tensor_operation/gpu/grid/gridwise_gemm_pipeline_v2.hpp"
#include "ck/tensor_operation/gpu/grid/gridwise_gemm_pipeline_v4_direct_load.hpp"

Go to the source code of this file.

Namespaces
	ck

Enumerations
enum class	ck::PipelineVersion {   ck::v1 ,   ck::v2 ,   ck::v4 ,   ck::weight_only }

Functions
template<PipelineVersion PipelineVer, index_t NumPrefetch = 1, LoopScheduler LoopSched = LoopScheduler::Default, bool AEnableLds = true, bool BEnableLds = true>
constexpr auto	ck::GridwiseGemmPipeline_Selector ()
std::ostream &	operator<< (std::ostream &os, const ck::PipelineVersion &p)

Function Documentation

◆ operator<<()

std::ostream& operator<< ( std::ostream &  os, const ck::PipelineVersion &  p  )

inline