ROCm/composable_kernel
1
0
Fork 0
mirror of https://github.com/ROCm/composable_kernel.git synced 2026-06-28 18:56:59 +00:00
Code Issues Packages Projects Releases Wiki Activity

Remove useless call of __builtin_amdgcn_s_waitcnt(0xc07f)

Browse Source
This commit is contained in:
Qianfeng Zhang
2025-12-17 07:35:19 +00:00
parent 97136bd3f7
commit e80c99b672
2 changed files with 0 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
example/ck_tile/18_hstu_attention/hstu_attention_no_softmax_fwd_pipeline.hpp
View File

@@ -336,8 +336,6 @@ struct HstuAttentionNoSoftmaxFwdPipelineQRKSVS
q_reg_tiles[i_rep] = load_tile(q_lds_read_window);
__builtin_amdgcn_s_waitcnt(0xc07f);
// the following codes will not generate actual instructions by the compiler
set_slice_tile(q_tile,
q_reg_tiles[i_rep],

2
example/ck_tile/18_hstu_attention/hstu_attention_with_softmax_fwd_pipeline.hpp
View File

@@ -353,8 +353,6 @@ struct HstuAttentionWithSoftmaxFwdPipelineQRKSVS
q_reg_tiles[i_rep] = load_tile(q_lds_read_window);
__builtin_amdgcn_s_waitcnt(0xc07f);
// the following codes will not generate actual instructions by the compiler
set_slice_tile(q_tile,
q_reg_tiles[i_rep],