Skip to content

[FA] add bwd variants for warp spec #91

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
wants to merge 9 commits into
base: main
Choose a base branch
from
Open

[FA] add bwd variants for warp spec #91

Show file tree
Hide file tree
Changes from 8 commits
Commits
Show all changes
9 commits
Select commit Hold shift + click to select a range
152d476
bwd two variants
manman-ren Nov 30, 2024
e652eb0
fix variants
manman-ren Dec 3, 2024
50db08c
make a copy for ws version
manman-ren Dec 3, 2024
8d2ca2e
use [1,2]
manman-ren Dec 3, 2024
bc63261
minor fix for original variant
manman-ren Dec 3, 2024
79c804f
skip tests
manman-ren Dec 3, 2024
757db43
lint
manman-ren Dec 3, 2024
2867e2f
small fix
manman-ren Dec 5, 2024
02ef2c2
fix fp8
manman-ren Dec 5, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
1 change: 1 addition & 0 deletions test/test_gpu/skip_tests_h100_pytorch.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -19,6 +19,7 @@ flash_attention:
- triton_tutorial_flash_v2_ws
- triton_tutorial_flash_v2_tma_ws
- triton_tutorial_flash_v2_tma_ws_persistent
- triton_tutorial_flash_v2_bwd_ws
fp8_attention:
- colfax_fmha
# triton_flash_v2 requires triton-main
Expand Down
1 change: 1 addition & 0 deletions test/test_gpu/skip_tests_h100_triton_main.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -11,6 +11,7 @@ flash_attention:
- triton_tutorial_flash_v2_ws
- triton_tutorial_flash_v2_tma_ws
- triton_tutorial_flash_v2_tma_ws_persistent
- triton_tutorial_flash_v2_bwd_ws
fp8_attention:
# fb-only kernel
- colfax_fmha
Expand Down
Loading
Loading