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
69224aac
Commit
69224aac
authored
Feb 23, 2023
by
fsx950223
Browse files
add grouped flash attention api
parent
7409bc5d
Changes
3
Expand all
Hide whitespace changes
Inline
Side-by-side
Showing
3 changed files
with
2084 additions
and
0 deletions
+2084
-0
example/32_batched_gemm_scale_softmax_gemm/CMakeLists.txt
example/32_batched_gemm_scale_softmax_gemm/CMakeLists.txt
+6
-0
example/32_batched_gemm_scale_softmax_gemm/grouped_multihead_attention_backward_fp16.cpp
...oftmax_gemm/grouped_multihead_attention_backward_fp16.cpp
+901
-0
include/ck/tensor_operation/gpu/device/impl/device_grouped_multihead_attention_backward_xdl_cshuffle_v2.hpp
..._grouped_multihead_attention_backward_xdl_cshuffle_v2.hpp
+1177
-0
No files found.
example/32_batched_gemm_scale_softmax_gemm/CMakeLists.txt
View file @
69224aac
...
...
@@ -5,7 +5,13 @@ add_example_executable(example_batched_gemm_scale_softmax_gemm_permute_xdl_bf16
add_example_executable
(
example_grouped_gemm_scale_softmax_gemm_permute_xdl_fp16 grouped_gemm_scale_softmax_gemm_permute_xdl_fp16.cpp
)
add_example_executable
(
example_batched_gemm_lower_triangle_scale_softmax_gemm_permute_xdl_fp16 batched_gemm_lower_triangle_scale_softmax_gemm_permute_xdl_fp16.cpp
)
add_example_executable
(
example_grouped_gemm_lower_triangle_scale_softmax_gemm_permute_xdl_fp16 grouped_gemm_lower_triangle_scale_softmax_gemm_permute_xdl_fp16.cpp
)
add_example_executable
(
example_grouped_multihead_attention_forward_fp16 grouped_multihead_attention_forward_fp16.cpp
)
add_example_executable
(
example_batched_multihead_attention_forward_fp16 batched_multihead_attention_forward_fp16.cpp
)
add_example_executable
(
example_grouped_multihead_attention_forward_bf16 grouped_multihead_attention_forward_bf16.cpp
)
add_example_executable
(
example_batched_multihead_attention_forward_bf16 batched_multihead_attention_forward_bf16.cpp
)
add_example_executable
(
example_batched_multihead_attention_backward_fp16 batched_multihead_attention_backward_fp16.cpp
)
add_example_executable
(
example_grouped_multihead_attention_backward_fp16 grouped_multihead_attention_backward_fp16.cpp
)
add_example_executable
(
example_batched_multihead_attention_backward_pt1_fp16 batched_multihead_attention_backward_pt1_fp16.cpp
)
add_custom_target
(
example_gemm_scale_softmax_gemm
)
add_dependencies
(
example_gemm_scale_softmax_gemm example_batched_gemm_scale_softmax_gemm_xdl_fp16
)
...
...
example/32_batched_gemm_scale_softmax_gemm/grouped_multihead_attention_backward_fp16.cpp
0 → 100644
View file @
69224aac
This diff is collapsed.
Click to expand it.
include/ck/tensor_operation/gpu/device/impl/device_grouped_multihead_attention_backward_xdl_cshuffle_v2.hpp
0 → 100644
View file @
69224aac
This diff is collapsed.
Click to expand it.
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