Skip to content
GitLab
Menu
Projects
Groups
Snippets
Loading...
Help
Help
Support
Community forum
Keyboard shortcuts
?
Submit feedback
Contribute to GitLab
Sign in / Register
Toggle navigation
Menu
Open sidebar
gaoqiong
composable_kernel
Commits
2f93e26f
Unverified
Commit
2f93e26f
authored
Nov 14, 2023
by
Dan Yao
Committed by
GitHub
Nov 14, 2023
Browse files
Merge pull request #1043 from ROCmSoftwarePlatform/mha-train-develop-tinyfix
Remove repeated definition in local scope
parents
f404e93a
a6497e17
Changes
2
Show whitespace changes
Inline
Side-by-side
Showing
2 changed files
with
0 additions
and
8 deletions
+0
-8
include/ck/tensor_operation/gpu/grid/gridwise_batched_mha_fwd_xdl_cshuffle_v2.hpp
...ion/gpu/grid/gridwise_batched_mha_fwd_xdl_cshuffle_v2.hpp
+0
-4
include/ck/tensor_operation/gpu/grid/gridwise_batched_mha_infer_xdl_cshuffle.hpp
...tion/gpu/grid/gridwise_batched_mha_infer_xdl_cshuffle.hpp
+0
-4
No files found.
include/ck/tensor_operation/gpu/grid/gridwise_batched_mha_fwd_xdl_cshuffle_v2.hpp
View file @
2f93e26f
...
...
@@ -1480,10 +1480,6 @@ struct GridwiseBatchedMultiheadAttentionForward_Xdl_CShuffle_V2
constexpr
index_t
MWave
=
MPerBlock
/
(
MXdlPerWave
*
MPerXdl
);
constexpr
index_t
NWave
=
Gemm1NPerBlock
/
(
Gemm1NXdlPerWave
*
NPerXdl
);
// TODO: hacky, fix it!
constexpr
auto
c_thread_desc_m0_n0_m1_n1_m2_n2_n3_n4
=
gemm1_blockwise_gemm
.
GetCThreadDescriptor_M0_N0_M1_N1_M2_N2_N3_N4
();
// TODO: hacky, fix it!
// c_block_desc_m0_n0_m1_n1_m2_n2_n3_n4_tmp is only used to get lengths
constexpr
auto
c_block_desc_m0_n0_m1_n1_m2_n2_n3_n4_tmp
=
...
...
include/ck/tensor_operation/gpu/grid/gridwise_batched_mha_infer_xdl_cshuffle.hpp
View file @
2f93e26f
...
...
@@ -1195,10 +1195,6 @@ struct GridwiseMultiHeadFlashAttentionInfer_Xdl_CShuffle
constexpr
index_t
MWave
=
MPerBlock
/
(
MXdlPerWave
*
MPerXdl
);
constexpr
index_t
NWave
=
Gemm1NPerBlock
/
(
Gemm1NXdlPerWave
*
NPerXdl
);
// TODO: hacky, fix it!
constexpr
auto
c_thread_desc_m0_n0_m1_n1_m2_n2_n3_n4
=
gemm1_blockwise_gemm
.
GetCThreadDescriptor_M0_N0_M1_N1_M2_N2_N3_N4
();
// TODO: hacky, fix it!
// c_block_desc_m0_n0_m1_n1_m2_n2_n3_n4_tmp is only used to get lengths
constexpr
auto
c_block_desc_m0_n0_m1_n1_m2_n2_n3_n4_tmp
=
...
...
Write
Preview
Markdown
is supported
0%
Try again
or
attach a new file
.
Attach a file
Cancel
You are about to add
0
people
to the discussion. Proceed with caution.
Finish editing this message first!
Cancel
Please
register
or
sign in
to comment