Skip to content

Fixes mask validation in forward kernel #57

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

Merged
merged 1 commit into from
Jul 3, 2025
Merged

Fixes mask validation in forward kernel #57

Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
3 changes: 1 addition & 2 deletions flash_dmattn/flash_dmattn_triton.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -132,8 +132,7 @@ def _fwd_kernel(
)

# Check if any element in mask is non-zero
# any_active = tl.sum(mask > 0) > 0
any_active = True
any_active = tl.sum(mask > 0) > 0
Copy link

Copilot AI Jul 3, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

[nitpick] Remove the commented-out legacy code now that the proper mask check is active to reduce clutter.

Copilot uses AI. Check for mistakes.

# compute acc_s
acc_s = tl.zeros([BLOCK_M, BLOCK_N], dtype=tl.float32)
Expand Down