biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

add cutlass support for blackwell fp8 gemm (#13798)

Browse Source
This commit is contained in:
kushanam
2025-03-04 07:55:07 -08:00
committed by GitHub
parent b3cf368d79
commit f89978ad7c
11 changed files with 272 additions and 65 deletions
Download Patch File Download Diff File Expand all files Collapse all files

25
csrc/quantization/cutlass_w8a8/scaled_mm_c3x.cu
View File

@@ -71,3 +71,28 @@ void cutlass_scaled_mm_azp_sm90(torch::Tensor& out, torch::Tensor const& a,
vllm::cutlass_scaled_mm_azp_sm90_int8(out, a, b, a_scales, b_scales, azp_adj,
azp, bias);
}
#if defined CUDA_VERSION && CUDA_VERSION >= 12080
void cutlass_scaled_mm_sm100(torch::Tensor& c, torch::Tensor const& a,
torch::Tensor const& b,
torch::Tensor const& a_scales,
torch::Tensor const& b_scales,
std::optional<torch::Tensor> const& bias) {
TORCH_CHECK(a_scales.dtype() == torch::kFloat32);
TORCH_CHECK(b_scales.dtype() == torch::kFloat32);
int M = a.size(0), N = b.size(1), K = a.size(1);
TORCH_CHECK(
(a_scales.numel() == 1 || a_scales.numel() == a.size(0)) &&
(b_scales.numel() == 1 || b_scales.numel() == b.size(1)),
"Currently, block scaled fp8 gemm is not implemented for Blackwell");
// Standard per-tensor/per-token/per-channel scaling
TORCH_CHECK(a_scales.is_contiguous() && b_scales.is_contiguous());
TORCH_CHECK(a.dtype() == torch::kFloat8_e4m3fn,
"Currently, only fp8 gemm is implemented for Blackwell");
vllm::cutlass_scaled_mm_sm100_fp8(c, a, b, a_scales, b_scales, bias);
}
#endif