Skip to content

[Bug] Fix Cutlass Scaled MM Compilation Error #24887

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 4 commits into from
Sep 15, 2025
Merged

[Bug] Fix Cutlass Scaled MM Compilation Error #24887

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
85e3b6e
fix cutlass compilation error
yewentao256 Sep 15, 2025
d86389d
update as gemini suggests
yewentao256 Sep 15, 2025
3612efa
fix CI
yewentao256 Sep 15, 2025
e674ef4
Merge branch 'main' into wye-fix-cutlass-compilation-error
yewentao256 Sep 15, 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
42 changes: 23 additions & 19 deletions csrc/quantization/cutlass_w8a8/c3x/scaled_mm_blockwise_sm100_fp8_dispatch.cuh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -146,6 +146,7 @@ void cutlass_gemm_caller_blockwise(torch::Tensor& out, torch::Tensor const& a,

using ElementAB = typename Gemm::ElementAB;
using ElementD = typename Gemm::ElementD;
using ElementBlockScale = typename Gemm::ElementBlockScale;

int32_t m = a.size(0), n = b.size(1), k = a.size(1);

Expand All @@ -166,26 +167,29 @@ void cutlass_gemm_caller_blockwise(torch::Tensor& out, torch::Tensor const& a,
ScaleConfig::tile_atom_to_shape_SFB(make_shape(n, m, k, 1)) :
ScaleConfig::tile_atom_to_shape_SFB(make_shape(m, n, k, 1));

auto a_ptr = static_cast<ElementAB*>(a.data_ptr());
auto b_ptr = static_cast<ElementAB*>(b.data_ptr());
auto a_scales_ptr = static_cast<float*>(a_scales.data_ptr());
auto b_scales_ptr = static_cast<float*>(b_scales.data_ptr());
auto a_ptr = static_cast<ElementAB const*>(a.data_ptr());
auto b_ptr = static_cast<ElementAB const*>(b.data_ptr());
auto a_scales_ptr = static_cast<ElementBlockScale const*>(a_scales.data_ptr());
auto b_scales_ptr = static_cast<ElementBlockScale const*>(b_scales.data_ptr());

auto mainloop_args = [&](){
// layout_SFA and layout_SFB cannot be swapped since they are deduced.
if (swap_ab) {
return typename GemmKernel::MainloopArguments{
b_ptr, b_stride, a_ptr, a_stride,
b_scales_ptr, layout_SFA, a_scales_ptr, layout_SFB
};
}
else {
return typename GemmKernel::MainloopArguments{
a_ptr, a_stride, b_ptr, b_stride,
a_scales_ptr, layout_SFA, b_scales_ptr, layout_SFB
};
}
}();
typename GemmKernel::MainloopArguments mainloop_args{};
mainloop_args.layout_SFA = layout_SFA;
mainloop_args.layout_SFB = layout_SFB;
if (swap_ab) {
mainloop_args.ptr_A = b_ptr;
mainloop_args.dA = b_stride;
mainloop_args.ptr_B = a_ptr;
mainloop_args.dB = a_stride;
mainloop_args.ptr_SFA = b_scales_ptr;
mainloop_args.ptr_SFB = a_scales_ptr;
} else {
mainloop_args.ptr_A = a_ptr;
mainloop_args.dA = a_stride;
mainloop_args.ptr_B = b_ptr;
mainloop_args.dB = b_stride;
mainloop_args.ptr_SFA = a_scales_ptr;
mainloop_args.ptr_SFB = b_scales_ptr;
}
auto prob_shape = swap_ab ? cute::make_shape(n, m, k, 1) : cute::make_shape(m, n, k, 1);

auto c_ptr = static_cast<ElementD*>(out.data_ptr());
Expand Down
26 changes: 15 additions & 11 deletions csrc/quantization/cutlass_w8a8/c3x/scaled_mm_blockwise_sm120_fp8_dispatch.cuh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -125,6 +125,7 @@ void cutlass_gemm_caller_blockwise(torch::Tensor& out, torch::Tensor const& a,

using ElementAB = typename Gemm::ElementAB;
using ElementD = typename Gemm::ElementD;
using ElementBlockScale = typename Gemm::ElementBlockScale;

int32_t m = a.size(0), n = b.size(1), k = a.size(1);

Expand All @@ -143,17 +144,20 @@ void cutlass_gemm_caller_blockwise(torch::Tensor& out, torch::Tensor const& a,
LayoutSFB layout_SFB =
ScaleConfig::tile_atom_to_shape_SFB(make_shape(m, n, k, 1));

auto a_ptr = static_cast<ElementAB*>(a.data_ptr());
auto b_ptr = static_cast<ElementAB*>(b.data_ptr());
auto a_scales_ptr = static_cast<float*>(a_scales.data_ptr());
auto b_scales_ptr = static_cast<float*>(b_scales.data_ptr());

auto mainloop_args = [&](){
return typename GemmKernel::MainloopArguments{
a_ptr, a_stride, b_ptr, b_stride,
a_scales_ptr, layout_SFA, b_scales_ptr, layout_SFB
};
}();
auto a_ptr = static_cast<ElementAB const*>(a.data_ptr());
auto b_ptr = static_cast<ElementAB const*>(b.data_ptr());
auto a_scales_ptr = static_cast<ElementBlockScale const*>(a_scales.data_ptr());
auto b_scales_ptr = static_cast<ElementBlockScale const*>(b_scales.data_ptr());

typename GemmKernel::MainloopArguments mainloop_args{};
mainloop_args.ptr_A = a_ptr;
mainloop_args.dA = a_stride;
mainloop_args.ptr_B = b_ptr;
mainloop_args.dB = b_stride;
mainloop_args.ptr_SFA = a_scales_ptr;
mainloop_args.layout_SFA = layout_SFA;
mainloop_args.ptr_SFB = b_scales_ptr;
mainloop_args.layout_SFB = layout_SFB;
auto prob_shape = cute::make_shape(m, n, k, 1);

auto c_ptr = static_cast<ElementD*>(out.data_ptr());
Expand Down
26 changes: 15 additions & 11 deletions csrc/quantization/cutlass_w8a8/c3x/scaled_mm_blockwise_sm90_fp8_dispatch.cuh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -115,6 +115,7 @@ void cutlass_gemm_caller_blockwise(torch::Tensor& out, torch::Tensor const& a,

using ElementAB = typename Gemm::ElementAB;
using ElementD = typename Gemm::ElementD;
using ElementBlockScale = typename Gemm::ElementBlockScale;

int32_t m = a.size(0), n = b.size(1), k = a.size(1);

Expand All @@ -135,17 +136,20 @@ void cutlass_gemm_caller_blockwise(torch::Tensor& out, torch::Tensor const& a,
LayoutSFB layout_SFB =
ScaleConfig::tile_atom_to_shape_SFB(make_shape(m, n, k, 1));

auto a_ptr = static_cast<ElementAB*>(a.data_ptr());
auto b_ptr = static_cast<ElementAB*>(b.data_ptr());
auto a_scales_ptr = static_cast<float*>(a_scales.data_ptr());
auto b_scales_ptr = static_cast<float*>(b_scales.data_ptr());

auto mainloop_args = [&](){
return typename GemmKernel::MainloopArguments{
a_ptr, a_stride, b_ptr, b_stride,
a_scales_ptr, layout_SFA, b_scales_ptr, layout_SFB
};
}();
auto a_ptr = static_cast<ElementAB const*>(a.data_ptr());
auto b_ptr = static_cast<ElementAB const*>(b.data_ptr());
auto a_scales_ptr = static_cast<ElementBlockScale const*>(a_scales.data_ptr());
auto b_scales_ptr = static_cast<ElementBlockScale const*>(b_scales.data_ptr());

typename GemmKernel::MainloopArguments mainloop_args{};
mainloop_args.ptr_A = a_ptr;
mainloop_args.dA = a_stride;
mainloop_args.ptr_B = b_ptr;
mainloop_args.dB = b_stride;
mainloop_args.ptr_SFA = a_scales_ptr;
mainloop_args.layout_SFA = layout_SFA;
mainloop_args.ptr_SFB = b_scales_ptr;
mainloop_args.layout_SFB = layout_SFB;
auto prob_shape = cute::make_shape(m, n, k, 1);

auto c_ptr = static_cast<ElementD*>(out.data_ptr());
Expand Down