Skip to content

Call fd.ops.sdpfa_fwd with kwargs #2799

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
KaelanDt merged 3 commits into from
Jan 3, 2026
+2 −7
Merged

Call fd.ops.sdpfa_fwd with kwargs #2799

Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
d42c6ee
Use kwargs
wujingyue Dec 13, 2025
c369b1d
Simplify
wujingyue Dec 13, 2025
4da05cc
Merge branch 'main' into wjy/sdpa
wujingyue Dec 29, 2025
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
9 changes: 2 additions & 7 deletions thunder/executors/nvfuserex_impl.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2568,13 +2568,8 @@ def _scaled_dot_product_flash_attention_forward(
fd: FusionDefinition,
lc_to_nv_map: dict,
) -> Any:
inputs = [query, key, value, dropout_p, is_causal, scale]
nv_inputs = []
for inp in inputs:
nv_inp = getnv(inp, fd, lc_to_nv_map) if inp is not None else None
nv_inputs.append(nv_inp)

return fd.ops.sdpfa_fwd(*nv_inputs)
args = [getnv(arg, fd, lc_to_nv_map) for arg in (query, key, value)]
return fd.ops.sdpfa_fwd(*args, dropout_p=dropoutp, is_causal=is_causal, scale=scale)


nv_sdpfa_fwd = ex.register_operator(
Expand Down
Loading