ROCm/composable_kernel
1
0
Fork 0
mirror of https://github.com/ROCm/composable_kernel.git synced 2026-05-19 12:30:16 +00:00
Code Issues Packages Projects Releases Wiki Activity

Modularize ckProfiler operations (#514)

Browse Source 
* 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]
This commit is contained in:
Po Yen Chen
2022-12-02 05:15:02 +08:00
committed by GitHub
parent 8e868bf880
commit 02db748e74
82 changed files with 346 additions and 273 deletions
Download Patch File Download Diff File Expand all files Collapse all files

10
profiler/src/profile_gemm_reduce.cpp
View File

@@ -6,7 +6,11 @@
#include <initializer_list>
#include <cstdlib>
#include "profiler/include/profile_gemm_reduce_impl.hpp"
#include "profiler/profile_gemm_reduce_impl.hpp"
#include "profiler_operation_registry.hpp"
#define OP_NAME "gemm_reduce"
#define OP_DESC "GEMM+Reduce"
int profile_gemm_reduce(int argc, char* argv[])
{
@@ -26,7 +30,7 @@ int profile_gemm_reduce(int argc, char* argv[])
if(!(argc == 14 || argc == 15))
{
printf("arg1: tensor operation (gemm: GEMM+Reduce)\n");
printf("arg1: tensor operation (" OP_NAME ": " OP_DESC ")\n");
printf("arg2: data type (0: fp32; 1: fp16)\n");
printf("arg3: matrix layout (0: A[m, k] * B[k, n] = C[m, n];\n");
printf(" 1: A[m, k] * B[n, k] = C[m, n];\n");
@@ -146,3 +150,5 @@ int profile_gemm_reduce(int argc, char* argv[])
return 0;
}
REGISTER_PROFILER_OPERATION(OP_NAME, OP_DESC, profile_gemm_reduce);