rocm-systems/hipamd/include/hip/amd_detail/amd_warp_functions.h

/*
Copyright (c) 2022 - 2023 Advanced Micro Devices, Inc. All rights reserved.

Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in
all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
THE SOFTWARE.
*/

#ifndef HIP_INCLUDE_HIP_AMD_DETAIL_WARP_FUNCTIONS_H
#define HIP_INCLUDE_HIP_AMD_DETAIL_WARP_FUNCTIONS_H

#if !defined(__HIPCC_RTC__)
#include "device_library_decls.h"  // ockl warp functions
#endif // !defined(__HIPCC_RTC__)

#if defined(__has_attribute) && __has_attribute(maybe_undef)
#define MAYBE_UNDEF __attribute__((maybe_undef))
#else
#define MAYBE_UNDEF
#endif

__device__ static inline unsigned __hip_ds_bpermute(int index, unsigned src) {
    union { int i; unsigned u; float f; } tmp; tmp.u = src;
    tmp.i = __builtin_amdgcn_ds_bpermute(index, tmp.i);
    return tmp.u;
}

__device__ static inline float __hip_ds_bpermutef(int index, float src) {
    union { int i; unsigned u; float f; } tmp; tmp.f = src;
    tmp.i = __builtin_amdgcn_ds_bpermute(index, tmp.i);
    return tmp.f;
}

__device__ static inline unsigned __hip_ds_permute(int index, unsigned src) {
    union { int i; unsigned u; float f; } tmp; tmp.u = src;
    tmp.i = __builtin_amdgcn_ds_permute(index, tmp.i);
    return tmp.u;
}

__device__ static inline float __hip_ds_permutef(int index, float src) {
    union { int i; unsigned u; float f; } tmp; tmp.f = src;
    tmp.i = __builtin_amdgcn_ds_permute(index, tmp.i);
    return tmp.f;
}

#define __hip_ds_swizzle(src, pattern)  __hip_ds_swizzle_N<(pattern)>((src))
#define __hip_ds_swizzlef(src, pattern) __hip_ds_swizzlef_N<(pattern)>((src))

template <int pattern>
__device__ static inline unsigned __hip_ds_swizzle_N(unsigned int src) {
    union { int i; unsigned u; float f; } tmp; tmp.u = src;
    tmp.i = __builtin_amdgcn_ds_swizzle(tmp.i, pattern);
    return tmp.u;
}

template <int pattern>
__device__ static inline float __hip_ds_swizzlef_N(float src) {
    union { int i; unsigned u; float f; } tmp; tmp.f = src;
    tmp.i = __builtin_amdgcn_ds_swizzle(tmp.i, pattern);
    return tmp.f;
}

#define __hip_move_dpp(src, dpp_ctrl, row_mask, bank_mask, bound_ctrl) \
  __hip_move_dpp_N<(dpp_ctrl), (row_mask), (bank_mask), (bound_ctrl)>((src))

template <int dpp_ctrl, int row_mask, int bank_mask, bool bound_ctrl>
__device__ static inline int __hip_move_dpp_N(int src) {
    return __builtin_amdgcn_mov_dpp(src, dpp_ctrl, row_mask, bank_mask,
                                    bound_ctrl);
}

#if defined(__SPIRV__)
    inline __device__ const struct final {
        __device__
        __attribute__((always_inline, const))
        operator int() const noexcept {
            return __builtin_amdgcn_wavefrontsize();
        }
    } warpSize{};
#else
    __device__
    static constexpr int warpSize = __AMDGCN_WAVEFRONT_SIZE;
#endif

// warp vote function __all __any __ballot
__device__
inline
int __all(int predicate) {
    return __ockl_wfall_i32(predicate);
}

__device__
inline
int __any(int predicate) {
    return __ockl_wfany_i32(predicate);
}

__device__
inline
unsigned long long int __ballot(int predicate) {
    return __builtin_amdgcn_ballot_w64(predicate);
}

__device__
inline
unsigned long long int __ballot64(int predicate) {
    return __ballot(predicate);
}

// See amd_warp_sync_functions.h for an explanation of this preprocessor flag.
#if !defined(HIP_DISABLE_WARP_SYNC_BUILTINS)
// Since threads in a wave do not make independent progress, __activemask()
// always returns the exact active mask, i.e, all active threads in the wave.
__device__
inline
unsigned long long __activemask() {
  return __ballot(true);
}
#endif // HIP_DISABLE_WARP_SYNC_BUILTINS

__device__ static inline unsigned int __lane_id() {
    if (warpSize == 32) return __builtin_amdgcn_mbcnt_lo(-1, 0);
    return  __builtin_amdgcn_mbcnt_hi(
        -1, __builtin_amdgcn_mbcnt_lo(-1, 0));
}

__device__
inline
int __shfl(MAYBE_UNDEF int var, int src_lane, int width = warpSize) {
    int self = __lane_id();
    int index = (src_lane & (width - 1)) + (self & ~(width-1));
    return __builtin_amdgcn_ds_bpermute(index<<2, var);
}
__device__
inline
unsigned int __shfl(MAYBE_UNDEF unsigned int var, int src_lane, int width = warpSize) {
     union { int i; unsigned u; float f; } tmp; tmp.u = var;
    tmp.i = __shfl(tmp.i, src_lane, width);
    return tmp.u;
}
__device__
inline
float __shfl(MAYBE_UNDEF float var, int src_lane, int width = warpSize) {
    union { int i; unsigned u; float f; } tmp; tmp.f = var;
    tmp.i = __shfl(tmp.i, src_lane, width);
    return tmp.f;
}
__device__
inline
double __shfl(MAYBE_UNDEF double var, int src_lane, int width = warpSize) {
    static_assert(sizeof(double) == 2 * sizeof(int), "");
    static_assert(sizeof(double) == sizeof(__hip_uint64_t), "");

    int tmp[2]; __builtin_memcpy(tmp, &var, sizeof(tmp));
    tmp[0] = __shfl(tmp[0], src_lane, width);
    tmp[1] = __shfl(tmp[1], src_lane, width);

    __hip_uint64_t tmp0 = (static_cast<__hip_uint64_t>(tmp[1]) << 32ull) | static_cast<__hip_uint32_t>(tmp[0]);
    double tmp1;  __builtin_memcpy(&tmp1, &tmp0, sizeof(tmp0));
    return tmp1;
}
__device__
inline
long __shfl(MAYBE_UNDEF long var, int src_lane, int width = warpSize)
{
    #ifndef _MSC_VER
    static_assert(sizeof(long) == 2 * sizeof(int), "");
    static_assert(sizeof(long) == sizeof(__hip_uint64_t), "");

    int tmp[2]; __builtin_memcpy(tmp, &var, sizeof(tmp));
    tmp[0] = __shfl(tmp[0], src_lane, width);
    tmp[1] = __shfl(tmp[1], src_lane, width);

    __hip_uint64_t tmp0 = (static_cast<__hip_uint64_t>(tmp[1]) << 32ull) | static_cast<__hip_uint32_t>(tmp[0]);
    long tmp1;  __builtin_memcpy(&tmp1, &tmp0, sizeof(tmp0));
    return tmp1;
    #else
    static_assert(sizeof(long) == sizeof(int), "");
    return static_cast<long>(__shfl(static_cast<int>(var), src_lane, width));
    #endif
}
__device__
inline
unsigned long __shfl(MAYBE_UNDEF unsigned long var, int src_lane, int width = warpSize) {
    #ifndef _MSC_VER
    static_assert(sizeof(unsigned long) == 2 * sizeof(unsigned int), "");
    static_assert(sizeof(unsigned long) == sizeof(__hip_uint64_t), "");

    unsigned int tmp[2]; __builtin_memcpy(tmp, &var, sizeof(tmp));
    tmp[0] = __shfl(tmp[0], src_lane, width);
    tmp[1] = __shfl(tmp[1], src_lane, width);

    __hip_uint64_t tmp0 = (static_cast<__hip_uint64_t>(tmp[1]) << 32ull) | static_cast<__hip_uint32_t>(tmp[0]);
    unsigned long tmp1;  __builtin_memcpy(&tmp1, &tmp0, sizeof(tmp0));
    return tmp1;
    #else
    static_assert(sizeof(unsigned long) == sizeof(unsigned int), "");
    return static_cast<unsigned long>(__shfl(static_cast<unsigned int>(var), src_lane, width));
    #endif
}
__device__
inline
long long __shfl(MAYBE_UNDEF long long var, int src_lane, int width = warpSize)
{
    static_assert(sizeof(long long) == 2 * sizeof(int), "");
    static_assert(sizeof(long long) == sizeof(__hip_uint64_t), "");

    int tmp[2]; __builtin_memcpy(tmp, &var, sizeof(tmp));
    tmp[0] = __shfl(tmp[0], src_lane, width);
    tmp[1] = __shfl(tmp[1], src_lane, width);

    __hip_uint64_t tmp0 = (static_cast<__hip_uint64_t>(tmp[1]) << 32ull) | static_cast<__hip_uint32_t>(tmp[0]);
    long long tmp1;  __builtin_memcpy(&tmp1, &tmp0, sizeof(tmp0));
    return tmp1;
}
__device__
inline
unsigned long long __shfl(MAYBE_UNDEF unsigned long long var, int src_lane, int width = warpSize) {
    static_assert(sizeof(unsigned long long) == 2 * sizeof(unsigned int), "");
    static_assert(sizeof(unsigned long long) == sizeof(__hip_uint64_t), "");

    unsigned int tmp[2]; __builtin_memcpy(tmp, &var, sizeof(tmp));
    tmp[0] = __shfl(tmp[0], src_lane, width);
    tmp[1] = __shfl(tmp[1], src_lane, width);

    __hip_uint64_t tmp0 = (static_cast<__hip_uint64_t>(tmp[1]) << 32ull) | static_cast<__hip_uint32_t>(tmp[0]);
    unsigned long long tmp1;  __builtin_memcpy(&tmp1, &tmp0, sizeof(tmp0));
    return tmp1;
}

__device__
inline
int __shfl_up(MAYBE_UNDEF int var, unsigned int lane_delta, int width = warpSize) {
    int self = __lane_id();
    int index = self - lane_delta;
    index = (index < (self & ~(width-1)))?self:index;
    return __builtin_amdgcn_ds_bpermute(index<<2, var);
}
__device__
inline
unsigned int __shfl_up(MAYBE_UNDEF unsigned int var, unsigned int lane_delta, int width = warpSize) {
    union { int i; unsigned u; float f; } tmp; tmp.u = var;
    tmp.i = __shfl_up(tmp.i, lane_delta, width);
    return tmp.u;
}
__device__
inline
float __shfl_up(MAYBE_UNDEF float var, unsigned int lane_delta, int width = warpSize) {
    union { int i; unsigned u; float f; } tmp; tmp.f = var;
    tmp.i = __shfl_up(tmp.i, lane_delta, width);
    return tmp.f;
}
__device__
inline
double __shfl_up(MAYBE_UNDEF double var, unsigned int lane_delta, int width = warpSize) {
    static_assert(sizeof(double) == 2 * sizeof(int), "");
    static_assert(sizeof(double) == sizeof(__hip_uint64_t), "");

    int tmp[2]; __builtin_memcpy(tmp, &var, sizeof(tmp));
    tmp[0] = __shfl_up(tmp[0], lane_delta, width);
    tmp[1] = __shfl_up(tmp[1], lane_delta, width);

    __hip_uint64_t tmp0 = (static_cast<__hip_uint64_t>(tmp[1]) << 32ull) | static_cast<__hip_uint32_t>(tmp[0]);
    double tmp1;  __builtin_memcpy(&tmp1, &tmp0, sizeof(tmp0));
    return tmp1;
}
__device__
inline
long __shfl_up(MAYBE_UNDEF long var, unsigned int lane_delta, int width = warpSize)
{
    #ifndef _MSC_VER
    static_assert(sizeof(long) == 2 * sizeof(int), "");
    static_assert(sizeof(long) == sizeof(__hip_uint64_t), "");

    int tmp[2]; __builtin_memcpy(tmp, &var, sizeof(tmp));
    tmp[0] = __shfl_up(tmp[0], lane_delta, width);
    tmp[1] = __shfl_up(tmp[1], lane_delta, width);

    __hip_uint64_t tmp0 = (static_cast<__hip_uint64_t>(tmp[1]) << 32ull) | static_cast<__hip_uint32_t>(tmp[0]);
    long tmp1;  __builtin_memcpy(&tmp1, &tmp0, sizeof(tmp0));
    return tmp1;
    #else
    static_assert(sizeof(long) == sizeof(int), "");
    return static_cast<long>(__shfl_up(static_cast<int>(var), lane_delta, width));
    #endif
}

__device__
inline
unsigned long __shfl_up(MAYBE_UNDEF unsigned long var, unsigned int lane_delta, int width = warpSize)
{
    #ifndef _MSC_VER
    static_assert(sizeof(unsigned long) == 2 * sizeof(unsigned int), "");
    static_assert(sizeof(unsigned long) == sizeof(__hip_uint64_t), "");

    unsigned int tmp[2]; __builtin_memcpy(tmp, &var, sizeof(tmp));
    tmp[0] = __shfl_up(tmp[0], lane_delta, width);
    tmp[1] = __shfl_up(tmp[1], lane_delta, width);

    __hip_uint64_t tmp0 = (static_cast<__hip_uint64_t>(tmp[1]) << 32ull) | static_cast<__hip_uint32_t>(tmp[0]);
    unsigned long tmp1;  __builtin_memcpy(&tmp1, &tmp0, sizeof(tmp0));
    return tmp1;
    #else
    static_assert(sizeof(unsigned long) == sizeof(unsigned int), "");
    return static_cast<unsigned long>(__shfl_up(static_cast<unsigned int>(var), lane_delta, width));
    #endif
}

__device__
inline
long long __shfl_up(MAYBE_UNDEF long long var, unsigned int lane_delta, int width = warpSize)
{
    static_assert(sizeof(long long) == 2 * sizeof(int), "");
    static_assert(sizeof(long long) == sizeof(__hip_uint64_t), "");
    int tmp[2]; __builtin_memcpy(tmp, &var, sizeof(tmp));
    tmp[0] = __shfl_up(tmp[0], lane_delta, width);
    tmp[1] = __shfl_up(tmp[1], lane_delta, width);
    __hip_uint64_t tmp0 = (static_cast<__hip_uint64_t>(tmp[1]) << 32ull) | static_cast<__hip_uint32_t>(tmp[0]);
    long long tmp1;  __builtin_memcpy(&tmp1, &tmp0, sizeof(tmp0));
    return tmp1;
}

__device__
inline
unsigned long long __shfl_up(MAYBE_UNDEF unsigned long long var, unsigned int lane_delta, int width = warpSize)
{
    static_assert(sizeof(unsigned long long) == 2 * sizeof(unsigned int), "");
    static_assert(sizeof(unsigned long long) == sizeof(__hip_uint64_t), "");
    unsigned int tmp[2]; __builtin_memcpy(tmp, &var, sizeof(tmp));
    tmp[0] = __shfl_up(tmp[0], lane_delta, width);
    tmp[1] = __shfl_up(tmp[1], lane_delta, width);
    __hip_uint64_t tmp0 = (static_cast<__hip_uint64_t>(tmp[1]) << 32ull) | static_cast<__hip_uint32_t>(tmp[0]);
    unsigned long long tmp1;  __builtin_memcpy(&tmp1, &tmp0, sizeof(tmp0));
    return tmp1;
}

__device__
inline
int __shfl_down(MAYBE_UNDEF int var, unsigned int lane_delta, int width = warpSize) {
    int self = __lane_id();
    int index = self + lane_delta;
    index = (int)((self&(width-1))+lane_delta) >= width?self:index;
    return __builtin_amdgcn_ds_bpermute(index<<2, var);
}
__device__
inline
unsigned int __shfl_down(MAYBE_UNDEF unsigned int var, unsigned int lane_delta, int width = warpSize) {
    union { int i; unsigned u; float f; } tmp; tmp.u = var;
    tmp.i = __shfl_down(tmp.i, lane_delta, width);
    return tmp.u;
}
__device__
inline
float __shfl_down(MAYBE_UNDEF float var, unsigned int lane_delta, int width = warpSize) {
    union { int i; unsigned u; float f; } tmp; tmp.f = var;
    tmp.i = __shfl_down(tmp.i, lane_delta, width);
    return tmp.f;
}
__device__
inline
double __shfl_down(MAYBE_UNDEF double var, unsigned int lane_delta, int width = warpSize) {
    static_assert(sizeof(double) == 2 * sizeof(int), "");
    static_assert(sizeof(double) == sizeof(__hip_uint64_t), "");

    int tmp[2]; __builtin_memcpy(tmp, &var, sizeof(tmp));
    tmp[0] = __shfl_down(tmp[0], lane_delta, width);
    tmp[1] = __shfl_down(tmp[1], lane_delta, width);

    __hip_uint64_t tmp0 = (static_cast<__hip_uint64_t>(tmp[1]) << 32ull) | static_cast<__hip_uint32_t>(tmp[0]);
    double tmp1;  __builtin_memcpy(&tmp1, &tmp0, sizeof(tmp0));
    return tmp1;
}
__device__
inline
long __shfl_down(MAYBE_UNDEF long var, unsigned int lane_delta, int width = warpSize)
{
    #ifndef _MSC_VER
    static_assert(sizeof(long) == 2 * sizeof(int), "");
    static_assert(sizeof(long) == sizeof(__hip_uint64_t), "");

    int tmp[2]; __builtin_memcpy(tmp, &var, sizeof(tmp));
    tmp[0] = __shfl_down(tmp[0], lane_delta, width);
    tmp[1] = __shfl_down(tmp[1], lane_delta, width);

    __hip_uint64_t tmp0 = (static_cast<__hip_uint64_t>(tmp[1]) << 32ull) | static_cast<__hip_uint32_t>(tmp[0]);
    long tmp1;  __builtin_memcpy(&tmp1, &tmp0, sizeof(tmp0));
    return tmp1;
    #else
    static_assert(sizeof(long) == sizeof(int), "");
    return static_cast<long>(__shfl_down(static_cast<int>(var), lane_delta, width));
    #endif
}
__device__
inline
unsigned long __shfl_down(MAYBE_UNDEF unsigned long var, unsigned int lane_delta, int width = warpSize)
{
    #ifndef _MSC_VER
    static_assert(sizeof(unsigned long) == 2 * sizeof(unsigned int), "");
    static_assert(sizeof(unsigned long) == sizeof(__hip_uint64_t), "");

    unsigned int tmp[2]; __builtin_memcpy(tmp, &var, sizeof(tmp));
    tmp[0] = __shfl_down(tmp[0], lane_delta, width);
    tmp[1] = __shfl_down(tmp[1], lane_delta, width);

    __hip_uint64_t tmp0 = (static_cast<__hip_uint64_t>(tmp[1]) << 32ull) | static_cast<__hip_uint32_t>(tmp[0]);
    unsigned long tmp1;  __builtin_memcpy(&tmp1, &tmp0, sizeof(tmp0));
    return tmp1;
    #else
    static_assert(sizeof(unsigned long) == sizeof(unsigned int), "");
    return static_cast<unsigned long>(__shfl_down(static_cast<unsigned int>(var), lane_delta, width));
    #endif
}
__device__
inline
long long __shfl_down(MAYBE_UNDEF long long var, unsigned int lane_delta, int width = warpSize)
{
    static_assert(sizeof(long long) == 2 * sizeof(int), "");
    static_assert(sizeof(long long) == sizeof(__hip_uint64_t), "");
    int tmp[2]; __builtin_memcpy(tmp, &var, sizeof(tmp));
    tmp[0] = __shfl_down(tmp[0], lane_delta, width);
    tmp[1] = __shfl_down(tmp[1], lane_delta, width);
    __hip_uint64_t tmp0 = (static_cast<__hip_uint64_t>(tmp[1]) << 32ull) | static_cast<__hip_uint32_t>(tmp[0]);
    long long tmp1;  __builtin_memcpy(&tmp1, &tmp0, sizeof(tmp0));
    return tmp1;
}
__device__
inline
unsigned long long __shfl_down(MAYBE_UNDEF unsigned long long var, unsigned int lane_delta, int width = warpSize)
{
    static_assert(sizeof(unsigned long long) == 2 * sizeof(unsigned int), "");
    static_assert(sizeof(unsigned long long) == sizeof(__hip_uint64_t), "");
    unsigned int tmp[2]; __builtin_memcpy(tmp, &var, sizeof(tmp));
    tmp[0] = __shfl_down(tmp[0], lane_delta, width);
    tmp[1] = __shfl_down(tmp[1], lane_delta, width);
    __hip_uint64_t tmp0 = (static_cast<__hip_uint64_t>(tmp[1]) << 32ull) | static_cast<__hip_uint32_t>(tmp[0]);
    unsigned long long tmp1;  __builtin_memcpy(&tmp1, &tmp0, sizeof(tmp0));
    return tmp1;
}

__device__
inline
int __shfl_xor(MAYBE_UNDEF int var, int lane_mask, int width = warpSize) {
    int self = __lane_id();
    int index = self^lane_mask;
    index = index >= ((self+width)&~(width-1))?self:index;
    return __builtin_amdgcn_ds_bpermute(index<<2, var);
}
__device__
inline
unsigned int __shfl_xor(MAYBE_UNDEF unsigned int var, int lane_mask, int width = warpSize) {
    union { int i; unsigned u; float f; } tmp; tmp.u = var;
    tmp.i = __shfl_xor(tmp.i, lane_mask, width);
    return tmp.u;
}
__device__
inline
float __shfl_xor(MAYBE_UNDEF float var, int lane_mask, int width = warpSize) {
    union { int i; unsigned u; float f; } tmp; tmp.f = var;
    tmp.i = __shfl_xor(tmp.i, lane_mask, width);
    return tmp.f;
}
__device__
inline
double __shfl_xor(MAYBE_UNDEF double var, int lane_mask, int width = warpSize) {
    static_assert(sizeof(double) == 2 * sizeof(int), "");
    static_assert(sizeof(double) == sizeof(__hip_uint64_t), "");

    int tmp[2]; __builtin_memcpy(tmp, &var, sizeof(tmp));
    tmp[0] = __shfl_xor(tmp[0], lane_mask, width);
    tmp[1] = __shfl_xor(tmp[1], lane_mask, width);

    __hip_uint64_t tmp0 = (static_cast<__hip_uint64_t>(tmp[1]) << 32ull) | static_cast<__hip_uint32_t>(tmp[0]);
    double tmp1;  __builtin_memcpy(&tmp1, &tmp0, sizeof(tmp0));
    return tmp1;
}
__device__
inline
long __shfl_xor(MAYBE_UNDEF long var, int lane_mask, int width = warpSize)
{
    #ifndef _MSC_VER
    static_assert(sizeof(long) == 2 * sizeof(int), "");
    static_assert(sizeof(long) == sizeof(__hip_uint64_t), "");

    int tmp[2]; __builtin_memcpy(tmp, &var, sizeof(tmp));
    tmp[0] = __shfl_xor(tmp[0], lane_mask, width);
    tmp[1] = __shfl_xor(tmp[1], lane_mask, width);

    __hip_uint64_t tmp0 = (static_cast<__hip_uint64_t>(tmp[1]) << 32ull) | static_cast<__hip_uint32_t>(tmp[0]);
    long tmp1;  __builtin_memcpy(&tmp1, &tmp0, sizeof(tmp0));
    return tmp1;
    #else
    static_assert(sizeof(long) == sizeof(int), "");
    return static_cast<long>(__shfl_xor(static_cast<int>(var), lane_mask, width));
    #endif
}
__device__
inline
unsigned long __shfl_xor(MAYBE_UNDEF unsigned long var, int lane_mask, int width = warpSize)
{
    #ifndef _MSC_VER
    static_assert(sizeof(unsigned long) == 2 * sizeof(unsigned int), "");
    static_assert(sizeof(unsigned long) == sizeof(__hip_uint64_t), "");

    unsigned int tmp[2]; __builtin_memcpy(tmp, &var, sizeof(tmp));
    tmp[0] = __shfl_xor(tmp[0], lane_mask, width);
    tmp[1] = __shfl_xor(tmp[1], lane_mask, width);

    __hip_uint64_t tmp0 = (static_cast<__hip_uint64_t>(tmp[1]) << 32ull) | static_cast<__hip_uint32_t>(tmp[0]);
    unsigned long tmp1;  __builtin_memcpy(&tmp1, &tmp0, sizeof(tmp0));
    return tmp1;
    #else
    static_assert(sizeof(unsigned long) == sizeof(unsigned int), "");
    return static_cast<unsigned long>(__shfl_xor(static_cast<unsigned int>(var), lane_mask, width));
    #endif
}
__device__
inline
long long __shfl_xor(MAYBE_UNDEF long long var, int lane_mask, int width = warpSize)
{
    static_assert(sizeof(long long) == 2 * sizeof(int), "");
    static_assert(sizeof(long long) == sizeof(__hip_uint64_t), "");
    int tmp[2]; __builtin_memcpy(tmp, &var, sizeof(tmp));
    tmp[0] = __shfl_xor(tmp[0], lane_mask, width);
    tmp[1] = __shfl_xor(tmp[1], lane_mask, width);
    __hip_uint64_t tmp0 = (static_cast<__hip_uint64_t>(tmp[1]) << 32ull) | static_cast<__hip_uint32_t>(tmp[0]);
    long long tmp1;  __builtin_memcpy(&tmp1, &tmp0, sizeof(tmp0));
    return tmp1;
}
__device__
inline
unsigned long long __shfl_xor(MAYBE_UNDEF unsigned long long var, int lane_mask, int width = warpSize)
{
    static_assert(sizeof(unsigned long long) == 2 * sizeof(unsigned int), "");
    static_assert(sizeof(unsigned long long) == sizeof(__hip_uint64_t), "");
    unsigned int tmp[2]; __builtin_memcpy(tmp, &var, sizeof(tmp));
    tmp[0] = __shfl_xor(tmp[0], lane_mask, width);
    tmp[1] = __shfl_xor(tmp[1], lane_mask, width);
    __hip_uint64_t tmp0 = (static_cast<__hip_uint64_t>(tmp[1]) << 32ull) | static_cast<__hip_uint32_t>(tmp[0]);
    unsigned long long tmp1;  __builtin_memcpy(&tmp1, &tmp0, sizeof(tmp0));
    return tmp1;
}

#endif