config.nvidia.hpp.in

#ifndef CK_CONFIG_NVIDIA_HPP
#define CK_CONFIG_NVIDIA_HPP

#include "cuda_runtime.h"
#include "cuda_fp16.h"
#include "nvToolsExt.h"
#include "helper_cuda.h"

// index type: unsigned or signed
#define CK_UNSIGNED_INDEX_TYPE 0

// device backend
#define CK_DEVICE_BACKEND_NVIDIA 1

// disable AMD inline asm and intrinsic
#define CK_USE_AMD_INLINE_ASM 0
#define CK_THREADWISE_GEMM_USE_AMD_INLINE_ASM 0
#define CK_USE_AMD_BUFFER_ADDRESSING 0
#define CK_USE_AMD_BUFFER_ADDRESSING_INTRINSIC 0
#define CK_USE_AMD_XDLOPS 0
#define CK_USE_AMD_XDLOPS_INLINE_ASM 0

// experimental implementation
#define CK_EXPERIMENTAL_BLOCKWISE_GEMM_USE_PIPELINE 0
#define CK_EXPERIMENTAL_TENSOR_COORDINATE_USE_CALCULATE_OFFSET_DIFF 0
#define CK_EXPERIMENTAL_THREADWISE_COPY_V4R2_USE_OPTIMIZED_ADDRESS_CACLULATION 0
#define CK_EXPERIMENTAL_USE_MORE_COMPILE_STATIC_BLOCKWISE_GENERIC_SLICE_COPY_V1 0
#define CK_EXPERIMENTAL_USE_MORE_COMPILE_STATIC_THREADWISE_GENERIC_TENSOR_SLICE_COPY_V1R2 0
#define CK_EXPERIMENTAL_USE_MORE_COMPILE_STATIC_THREADWISE_GENERIC_TENSOR_SLICE_COPY_V2R1 0

namespace ck {

enum AddressSpace
{
    generic,
    global,
    lds,
    vgpr
};

enum InMemoryDataOperation
{
    none,
    atomic_add
};

#if CK_UNSIGNED_INDEX_TYPE
using index_t = uint32_t;
#else
using index_t = int32_t;
#endif

} // namespace ck
#endif