mirror of
https://github.com/ROCm/composable_kernel.git
synced 2026-05-14 10:09:41 +00:00
* Re-structure ckProfiler source files
* Rename profiler.cpp to main.cpp
* Modularize ckProfiler operations
* Add description for profiler operations
* Use longer name to avoid name collision
* Use macro to delay expansion
* Use std::move() to avoid object copying
* Prohibit users from calling dtor
* Use macro to eliminate redundant code
* Make friend function hidden
* Add missing include directive <iostream>
* Fix wrong include directives
* Remove int8 from batchnorm-forward instances since it is not needed for forward training and could fail test
Co-authored-by: Qianfeng Zhang <Qianfeng.Zhang@amd.com>
[ROCm/composable_kernel commit: 8784a72e23]
50 lines
1.4 KiB
C++
50 lines
1.4 KiB
C++
// SPDX-License-Identifier: MIT
|
|
// Copyright (c) 2018-2022, Advanced Micro Devices, Inc. All rights reserved.
|
|
|
|
#include <iostream>
|
|
|
|
#include "profiler/profile_gemm_reduce_impl.hpp"
|
|
|
|
int main()
|
|
{
|
|
using Row = ck::tensor_layout::gemm::RowMajor;
|
|
using Col = ck::tensor_layout::gemm::ColumnMajor;
|
|
|
|
int M = 512;
|
|
int N = 256;
|
|
int K = 128;
|
|
|
|
bool pass = true;
|
|
|
|
pass = pass &&
|
|
ck::profiler::
|
|
profile_gemm_reduce_impl<ck::half_t, ck::half_t, ck::half_t, float, Row, Row, Row>(
|
|
true, 1, false, false, M, N, K, K, N, N);
|
|
|
|
pass = pass &&
|
|
ck::profiler::
|
|
profile_gemm_reduce_impl<ck::half_t, ck::half_t, ck::half_t, float, Row, Col, Row>(
|
|
true, 1, false, false, M, N, K, K, K, N);
|
|
|
|
pass = pass &&
|
|
ck::profiler::
|
|
profile_gemm_reduce_impl<ck::half_t, ck::half_t, ck::half_t, float, Col, Row, Row>(
|
|
true, 1, false, false, M, N, K, M, N, N);
|
|
|
|
pass = pass &&
|
|
ck::profiler::
|
|
profile_gemm_reduce_impl<ck::half_t, ck::half_t, ck::half_t, float, Col, Col, Row>(
|
|
true, 1, false, false, M, N, K, M, K, N);
|
|
|
|
if(pass)
|
|
{
|
|
std::cout << "test GEMM+Reduce fp16: Pass" << std::endl;
|
|
return 0;
|
|
}
|
|
else
|
|
{
|
|
std::cout << "test GEMM+Reduce fp16: Fail" << std::endl;
|
|
return -1;
|
|
}
|
|
}
|