ROCm/composable_kernel
1
0
Fork 0
mirror of https://github.com/ROCm/composable_kernel.git synced 2026-07-01 04:07:56 +00:00
Code Issues Packages Projects Releases Wiki Activity

fix readme

Browse Source
This commit is contained in:
Yanxing-Shi
2025-05-26 06:45:36 +00:00
parent 68fe3876e4
commit 7549e2b2e6
2 changed files with 6 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
tile_engine/ops/gemm/README.md
View File

@@ -36,6 +36,8 @@ rm -rf tile_engine/ && ninja benchmark_gemm # rebuild
-stride_b The stride value for tensor B. Default is 0.
-stride_c The stride value for tensor C Default is 0.
-split_k The split value for k dimension. Default is 1.
-enable_profile_cache Whether flush profile cache or not when benchmark kernel. Possible values are or false. Default is true.
-flush_profile_cache Whether flush profile cache or not when benchmark kernel. Possible values are true or false. Default is false.
-v The type of validation. Set to 0 for no validation, 1 for validation on CPU, or 2 for validation on GPU. Default is 2, validation on GPU.
-log Wether output kernel instance information or not. Possible values are true or false. Default is false.
-warmup The number of iterations before benchmark the kernel. Default is 50.

8
tile_engine/ops/gemm/gemm_host_api.hpp
View File

@@ -75,12 +75,12 @@ inline auto create_args(int argc, char* argv[])
.insert("split_k", "1", "The split value for k dimension. Default is 1.")
.insert("enable_profile_cache",
"true",
"whether use profile cache or not when benchmark kernel, Possible values are true "
"or false. Default is true")
"Whether use profile cache or not when benchmark kernel, Possible values are true "
"or false. Default is true.")
.insert("flush_profile_cache",
"false",
"whether flush profile cache or not when benchmark kernel. Possible values are "
"true or false. Default is false")
"Whether flush profile cache or not when benchmark kernel. Possible values are "
"true or false. Default is false.")
.insert("verify",
"2",
"The type of validation. Set to 0 for no validation, 1 for validation on CPU, or 2 "