Skip to content

Fix Wq size check #3228

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
Oct 23, 2025
Merged

Fix Wq size check #3228

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: 2 additions & 1 deletion ...o/csrc/cuda/rowwise_scaled_linear_sparse_cutlass/rowwise_scaled_linear_sparse_cutlass.cuh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -33,6 +33,7 @@
#endif

#define OPERATOR_NAME "rowwise_scaled_linear_sparse_cutlass"
#define PAD_TO_MULTIPLE_OF_16(x) (((x) + 15) / 16 * 16)

namespace torchao {

Expand Down Expand Up @@ -448,7 +449,7 @@ check_inputs(
// W_meta may be padded, thus expected shape calculations for this
// tensor are as follows.
const auto W_meta_size_0_expected = std::max((int)Wq_sizes[0], 64);
const auto W_meta_size_1_expected = std::max((int)Wq_sizes[1] / 4, 16);
const auto W_meta_size_1_expected = std::max(PAD_TO_MULTIPLE_OF_16((int)Wq_sizes[1]/4), 16);
TORCH_CHECK(W_meta.size(0) == W_meta_size_0_expected, OPERATOR_NAME,
" : Expected Wq meta argument to have ", W_meta_size_0_expected,
" rows, got ", W_meta.size(0), " rows");
Expand Down
Loading