Skip to content

[Flex Attention] Add a Triton version so we can test autoWS #983

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

Draft
manman-ren wants to merge 7 commits into
base: main
Choose a base branch
from
Draft

[Flex Attention] Add a Triton version so we can test autoWS #983

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
7 commits
Select commit Hold shift + click to select a range
3bf40f7
flex attention
manman-ren Mar 29, 2026
c2fa575
try merging forOps
manman-ren Mar 30, 2026
aada203
a single forOp
manman-ren Mar 30, 2026
c2f88c2
remove if branch to enable autoWS
manman-ren Mar 31, 2026
1ba04b5
add TMA/ws/persistent
manman-ren Mar 31, 2026
a84a90d
Enable autoWS for flex attention
manman-ren Apr 11, 2026
cae83de
dp 2 fail accuracy
manman-ren Apr 13, 2026
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
10 changes: 5 additions & 5 deletions tritonbench/kernels/blackwell_triton_fused_attention.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -276,14 +276,14 @@ def make_standard_config(BM, BN, s, w, subtile, vectmul, add2reduce, maxreg):

configs = [
make_standard_config(BM, BN, s, w, subtile, vectmul, add2reduce, maxreg)
for BM in [128, 256]
for BN in [64, 128]
for s in NUM_STAGES_OPTIONS
for BM in [256] # 128, 256]
for BN in [128] # 64, 128]
for s in [3] # NUM_STAGES_OPTIONS
for w in [4]
for subtile in [True]
for vectmul in [0, 1]
for vectmul in [1] # 0, 1]
for add2reduce in [False]
for maxreg in [152, 192]
for maxreg in [152] # , 192]
]


Expand Down
Loading
Loading