Skip to content

[CI Failure] Fix test_flashinfer_cutlass_mxfp4_mxfp8_fused_moe #24750

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 3 commits into from
Sep 13, 2025
Merged

[CI Failure] Fix test_flashinfer_cutlass_mxfp4_mxfp8_fused_moe #24750

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
e776c41
Fix test_flashinfer_cutlass_mxfp4_mxfp8_fused_moe
mgoin Sep 12, 2025
2243787
Merge branch 'main' into fix-blackwell-mxfp4-test
mgoin Sep 12, 2025
f84512c
Fix dummy def for sm100_cutlass_mla_decode
mgoin Sep 12, 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
1 change: 1 addition & 0 deletions csrc/attention/mla/sm100_cutlass_mla_kernel.cu
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -43,6 +43,7 @@ void sm100_cutlass_mla_decode(
torch::Tensor const& seq_lens,
torch::Tensor const& page_table,
torch::Tensor const& workspace,
double sm_scale,
int64_t num_kv_splits) {
TORCH_CHECK(false, "CUDA version must be >= 12.4 for cutlass_mla_decode");
}
Expand Down
4 changes: 2 additions & 2 deletions tests/kernels/moe/test_mxfp4_moe.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -771,11 +771,11 @@ def dequant_mxfp4_batches(mat_fp4: torch.Tensor,
w13_ref = dequant_mxfp4_batches(
w13_q.view(torch.uint8),
w13_scale.view(torch.uint8).reshape(-1)).to(torch.float32).reshape(
num_experts, 2 * intermediate_size, hidden_size)
num_experts, 2 * intermediate_size, hidden_size).to(device)
w2_ref = dequant_mxfp4_batches(
w2_q.view(torch.uint8),
w2_scale.view(torch.uint8).reshape(-1)).to(torch.float32).reshape(
num_experts, hidden_size, intermediate_size)
num_experts, hidden_size, intermediate_size).to(device)
Copy link
Collaborator

Choose a reason for hiding this comment

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

What will happen if we don't add the to(device) here?


# Quantize activations for SM100 path and dequantize for reference
hidden_states_q, hidden_states_sf = mxfp8_quantize(hidden_states, True, 32)
Expand Down